<%BANNER%>

Pharmacokinetics/Pharmacodynamics of Voriconazole

Permanent Link: http://ufdc.ufl.edu/UFE0022512/00001

Material Information

Title: Pharmacokinetics/Pharmacodynamics of Voriconazole
Physical Description: 1 online resource (155 p.)
Language: english
Creator: Li, Yanjun
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: candida, curve, dynamic, hplc, kill, microdialysis, pharmacodynamics, pharmacokinetics, time, voriconazole
Pharmacy -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Candida spp. is the most common cause of opportunistic mycoses worldwide, especially the invasive opportunistic mycotic infection. Oropharyngeal candidiasis is the most common opportunistic infection associated with AIDS. Candidemia is the fourth most common bloodstream infection in the developed world and is associated with mortality rates of up to 38% despite treatment with available antifungals. Voriconazole, a triazole agent that is effective in treating candidiasis and aspergillosis, inhibits ergosterol synthesis by blocking the action of 14alpha-demethylase. It has also lowered toxicity compared with previous antifungal agents, however, it is used to cure candidiasis inside the human body after oral administration or i.v. injection. Empiric antifungal therapy is initiated as the first measure dealing with fungal infection, and there are no suitable mathematical models of voriconazole antifungal activity established for rational guidance to maximize efficiency and minimize toxicity. The aim of these studies was to develop an optimal pharmacokinetic/pharmacodynamic (PK/PD) model to predict clinical outcome and improve the effectiveness of voriconazole. The PK/PD model integrates the pharmacokinetics of unbound voriconazole with in vitro antifungal activity against certain Candida strains for rational dosing of this agent. In these studies, we measured voriconazole activity against Candida isolates using time-kill methods validated by high performance liquid chromatography in vitro and a pharmacokinetic/pharmacodynamic mathematical model based on time-kill curves against different Candida strains was used to accurately describe the antifungal activity of voriconazole. Using this model, pharmacodynamic studies in dynamic system for describing the activity of voriconazole against Candida spp. were conducted. Moreover, in vitro microdialysis was performed to determine protein binding of voriconazole in both rat and human plasma. Also, in vivo microdialysis in Wistar rats after intravenous administration of voriconazole was conducted to evaluate unbound muscle concentrations of voriconazole and to compare the free tissue concentration to free plasma concentration. We investigated the voriconazole PK profile by analyzing the total concentrations in plasma and unbound concentrations in muscle and developed a population PK model to fit the voriconazole PK data in rats using NONMEM. These studies provided new insights to improve the drug clinical efficacy by using the developed voriconazole PK/PD model.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yanjun Li.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Derendorf, Hartmut C.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022512:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022512/00001

Material Information

Title: Pharmacokinetics/Pharmacodynamics of Voriconazole
Physical Description: 1 online resource (155 p.)
Language: english
Creator: Li, Yanjun
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: candida, curve, dynamic, hplc, kill, microdialysis, pharmacodynamics, pharmacokinetics, time, voriconazole
Pharmacy -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Candida spp. is the most common cause of opportunistic mycoses worldwide, especially the invasive opportunistic mycotic infection. Oropharyngeal candidiasis is the most common opportunistic infection associated with AIDS. Candidemia is the fourth most common bloodstream infection in the developed world and is associated with mortality rates of up to 38% despite treatment with available antifungals. Voriconazole, a triazole agent that is effective in treating candidiasis and aspergillosis, inhibits ergosterol synthesis by blocking the action of 14alpha-demethylase. It has also lowered toxicity compared with previous antifungal agents, however, it is used to cure candidiasis inside the human body after oral administration or i.v. injection. Empiric antifungal therapy is initiated as the first measure dealing with fungal infection, and there are no suitable mathematical models of voriconazole antifungal activity established for rational guidance to maximize efficiency and minimize toxicity. The aim of these studies was to develop an optimal pharmacokinetic/pharmacodynamic (PK/PD) model to predict clinical outcome and improve the effectiveness of voriconazole. The PK/PD model integrates the pharmacokinetics of unbound voriconazole with in vitro antifungal activity against certain Candida strains for rational dosing of this agent. In these studies, we measured voriconazole activity against Candida isolates using time-kill methods validated by high performance liquid chromatography in vitro and a pharmacokinetic/pharmacodynamic mathematical model based on time-kill curves against different Candida strains was used to accurately describe the antifungal activity of voriconazole. Using this model, pharmacodynamic studies in dynamic system for describing the activity of voriconazole against Candida spp. were conducted. Moreover, in vitro microdialysis was performed to determine protein binding of voriconazole in both rat and human plasma. Also, in vivo microdialysis in Wistar rats after intravenous administration of voriconazole was conducted to evaluate unbound muscle concentrations of voriconazole and to compare the free tissue concentration to free plasma concentration. We investigated the voriconazole PK profile by analyzing the total concentrations in plasma and unbound concentrations in muscle and developed a population PK model to fit the voriconazole PK data in rats using NONMEM. These studies provided new insights to improve the drug clinical efficacy by using the developed voriconazole PK/PD model.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yanjun Li.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Derendorf, Hartmut C.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022512:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101220_AAAAJI INGEST_TIME 2010-12-21T03:41:58Z PACKAGE UFE0022512_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 32207 DFID F20101220_AACRZY ORIGIN DEPOSITOR PATH li_y_Page_109.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
b60b9d7f270fd18d3b49330013a10e1f
SHA-1
f91a4488596a70e598bcb65c094ec9f6fca44498
114805 F20101220_AACRCR li_y_Page_020.jpg
4702b5197ee9c5de7734713051cf4470
f3af7b50276603cac73071ebad0170acae81ee54
3267 F20101220_AACQXL li_y_Page_043thm.jpg
69898bc46d894801e16695b5fa0053cc
07498bb1cb4cc7013907b20e5fd96cbd67262bdd
83081 F20101220_AACQWW li_y_Page_100.jpg
8c0776b982a26cbe5d49cf6a8c616409
28cbb76394c438ea52ba34c4e6ed3c7a8445d7e4
4395 F20101220_AACRDF li_y_Page_090thm.jpg
4da007ad8da0abc7a1d10b05a21174e9
8fcd7667a64674dc5d4c5614a3142c160a40a819
1725 F20101220_AACRZZ li_y_Page_112thm.jpg
7a34d8f3eea59440b036a12a7d84fdb0
2c478e255e2b7f8b30d04ec1927d06ad574ab368
88735 F20101220_AACQYA li_y_Page_053.jpg
c562eb8207a6ccda7ebe432870c813a0
0dc493fd32fe6dc5598fc2fb63cffa3f367c213a
18714 F20101220_AACRCS li_y_Page_125.QC.jpg
237de06bda4cf920566263821c137eb7
f5d42d7665fc373caed17d3b9da5336f95de7d52
831 F20101220_AACQWX li_y_Page_122.txt
9af4214d841615da3ffbc537edbcd79d
fcedaa19b7d4fd1ab6cf37fa9a6a18a11867a904
52151 F20101220_AACRDG li_y_Page_017.pro
4c8e9107dbf3b7f837069a0a35254789
5e6c030c6e49759659127dd5e15e55ca635bcbcf
25271604 F20101220_AACQYB li_y_Page_051.tif
75e53f57ce74c85c035a63aef652011a
2356b4051d1b58b648e62b81bd9f451f35a3e829
103777 F20101220_AACRCT li_y_Page_015.jpg
f5baf77f5a82961523dfbc9fecdd0705
ce8318c22238847cebbe2a3e6173922d73bcae24
66645 F20101220_AACQXM li_y_Page_009.pro
13046693147e214b92415d8693942688
c8052b52cf6a1e167f5e0771bed014ce9ad061f0
13870 F20101220_AACQWY li_y_Page_058.QC.jpg
5e7bdcbd44fc1a0ca08e2a985f95d0ab
2fd18b1f4e1fd0c64c88112482cb6e46190852e4
26503 F20101220_AACRDH li_y_Page_004.QC.jpg
870a739008c85b71b3001123b3334831
6c9c1a175a7cfa5249791d7d34e8e2687868b09a
13270 F20101220_AACQYC li_y_Page_120.QC.jpg
c34ff7b3c76542defbf5f125fd1ff424
ef7574f423bc2cfb678c0913fd7a54d981a01e39
F20101220_AACRCU li_y_Page_060.tif
8352f1faa738b1c52e5d44032bd86b95
b7f3992478960b094b12381e2f0192b26939292f
47186 F20101220_AACQXN li_y_Page_110.pro
8441da4d266e925098f5b84538204711
16de1a23fbca6126a0b7736929c563eecb70ca9f
19920 F20101220_AACQWZ li_y_Page_099.QC.jpg
4e75ed1d5105604b5fd82a231a9f02b2
0673e987e1540c0b006752fe445cf689b0da4880
982328 F20101220_AACRDI li_y_Page_131.jp2
95d84118f504917a23c92639fbbd64cf
4e94d576ce2d3cccd59bb794ec40455c20f64cac
1037 F20101220_AACQYD li_y_Page_059.txt
b841944677eb82b774a0e11a5574b711
2d04d89b3438feb8b77d9daefb40c0fa7ecc9c22
21141 F20101220_AACRCV li_y_Page_002.jp2
baf7ddd909d6bdb7ac7d5c662ada69bb
39dd4bb96f4fec32b9f3ec094783aa50af71c703
1730 F20101220_AACQXO li_y_Page_052.txt
cdbef4b520ef87925f12369566adbb3c
11f324730eff9e5fa87791e70fa030931677a9e0
F20101220_AACRDJ li_y_Page_029.tif
020ec411abbe07a11f5a12b7c9b2afbe
7d96916e796f6aa1dca7b120122a2e1a6a0434f1
657950 F20101220_AACQYE li_y_Page_099.jp2
7f7b8ec08158b0bc51cc8c40c00902ac
eb1defb6a413684e5780b6be3dfa398a2c2660d2
25265604 F20101220_AACQXP li_y_Page_071.tif
22e1ce749f85d1d574409540351f4444
155b86fb2485aeb846c92839cfc6d0c208ec4fff
17284 F20101220_AACRDK li_y_Page_088.pro
06542e50c3e6151fd550df3437ded655
b297204e5d1901122b847b843b76125dde1798b7
104630 F20101220_AACQYF li_y_Page_067.jpg
1c6ee3c07d2be8de083d55122281a741
3b18a3fb41d5a960d710cbe828ecc716d106e781
2133 F20101220_AACRCW li_y_Page_138.txt
e1fff7924975ec1ddbdd525f69bfe464
d458c422a759e297c282d9e11c62957840a36ce5
1004235 F20101220_AACQXQ li_y_Page_094.jp2
5b233cfd37f8ca051a3d028412c6c425
fe56c3785a30aa9df51ecae685f892953c8e52cc
35000 F20101220_AACRDL li_y_Page_019.QC.jpg
d7cbb658c3edc248835549a76f4206fb
db21ef3021485bc31c1558c51e5d326512933329
8609 F20101220_AACQYG li_y_Page_070thm.jpg
1a06efb55eee84c77c571e596554c8ca
a5ca1ab984bf6474db7b5465d1a2bdd83f21322b
1323 F20101220_AACRCX li_y_Page_076.txt
365bd86312b60cea4aa53aee364a31dc
38c6e1c4b23b0ed71d70b6187c816604fc3f136f
37897 F20101220_AACQXR li_y_Page_009.QC.jpg
05ff50258eaf015d140a6c33ae3aee71
7027e9fa2aade5d92fa44abe5e06d644a1bfa396
2296 F20101220_AACREA li_y_Page_023.txt
de3e398f88f6a6c2d30c710214ca69bc
32f97c6caff57491850f092cabc3cc8e16a0ca29
F20101220_AACRDM li_y_Page_007.tif
f4a722ec3516289b6691df691686ac93
211e5b8b40abf445a9bdca0627478f5e515b1150
10194 F20101220_AACQYH li_y_Page_057.QC.jpg
1814bb7bd7c63ae26ae4e19d9cbeaef3
637b61734485eb6fa93515e6fb2683da3b73cd10
86826 F20101220_AACRCY li_y_Page_092.jpg
ac25adeaf830f59a400ac7e35a0ac88f
e1a23d367c59ecea0e099d0fb84d7847f2a4942b
7625 F20101220_AACQXS li_y_Page_102thm.jpg
ca9c3a136784cef8970580e8b33695b4
1f170e66d788ed8001fc8512f1633e9659f5509e
1411 F20101220_AACREB li_y_Page_154thm.jpg
cd894c58d92b8ff9db293c393829c8a2
64153706509044e2494844a35acfab4236056894
7195 F20101220_AACRDN li_y_Page_095thm.jpg
bc759d00082073088d75f2307dd52b10
9516e4658335ed923c38594ab532ee30a852c2ee
59806 F20101220_AACQYI li_y_Page_129.jpg
cf976ad26610d1c42cd25cd0b495792b
7203dc88c1c28c4520a02f4111817605c8c1e4a5
102499 F20101220_AACRCZ li_y_Page_065.jpg
e17574b7078dabc3d563242737d4ffd9
63a4afcf04b2bad4a41ae0be2fd542cbe7e6eb8c
8704 F20101220_AACQXT li_y_Page_010thm.jpg
44ac97a7465938a652e551787bf2b402
f7abc63d29e15c1bc9090d8531476267f3c92895
17685 F20101220_AACREC li_y_Page_058.pro
cf0314d030dcf3f13869c8b068435eb9
e2fe3ac03f35861a9ddaa83e7330d2b2a88132d7
2079 F20101220_AACRDO li_y_Page_139.txt
4df264e4438c4a7575ed3985dc106a61
665e632a55d0119617eff71dc459b28f845a38ad
7753 F20101220_AACQYJ li_y_Page_094thm.jpg
2471f880a86b5f71ee37d86ac72b520f
fa81f9bccda49a708d983f2b6c644a763ebd8131
94914 F20101220_AACQXU li_y_Page_106.jpg
d2b4cdf50089c9d6e49735084ef71bed
2370ecc0e114da9c4520ac769af3367ce79835fe
15783 F20101220_AACRED li_y_Page_064.QC.jpg
c8e04137627cee4cbf1ad2547df68d57
2ea94a8aa7ff679d63de5231e7604df0c315b393
1446 F20101220_AACRDP li_y_Page_113.txt
9f8c8f2002a4a04e88f2f31c7413a262
0c7e8b820b4a1bbda337ec9580fc889a97a71ba2
F20101220_AACQYK li_y_Page_008.tif
f6fedecf5a37ce0b0869c10c81f516ac
43750649d856189c9893613d2472d8091e7e1d5c
21140 F20101220_AACQXV li_y_Page_154.jpg
db0ebe1c03fb2a5111f3b873fdd3e930
40965493c5ab1e43bc56e4085eeac7b3b3572a9f
1270 F20101220_AACREE li_y_Page_117.txt
d309240edd7448c3e9f160dccb8d223c
a73ab70c8aa9e9c3b26f25bee708bd622f9cb66a
33410 F20101220_AACRDQ li_y_Page_032.QC.jpg
91b22adeea3f8e35d6787c5e07fd4a03
5b3af22f4e1439f98dd6ea2bfa1041ad892b25aa
F20101220_AACQYL li_y_Page_070.tif
344485200ace72a0a01a9cfb8c796db6
add01de689bb5c2bf34e192cceaeb08a498ab55c
39208 F20101220_AACQXW li_y_Page_004.pro
a9f3546b004de7301293b546a2de6da0
3c1cc00d3e040ada953407015028d885cce55e95
F20101220_AACREF li_y_Page_137.tif
21ddd4904390ebb6376a8d661f37194f
e1fc2f8c39c28efa64896f05c9767d6155cd779c
8906 F20101220_AACRDR li_y_Page_141thm.jpg
4b2b22224bdb78fe0afad7af1ea949d8
6501e304e40af0366bc8b5bef2f1e5f2d9f2b43a
37217 F20101220_AACQYM li_y_Page_147.QC.jpg
81f31f7571ed51d103b07eefa82df2f1
406ef2cf720d877c422e919d86a592b246808a52
40259 F20101220_AACQXX li_y_Page_088.jpg
bc7e616a117261389580f6c6459fb253
a3ed299e27ec813e7ceef1ce38e927625d876377
1042217 F20101220_AACREG li_y_Page_110.jp2
9f636826fa3cec758de0dfa5051ed1c3
40c778435df179c2f5206a6890faab2e5989abef
1106 F20101220_AACQZA li_y_Page_056.txt
2e9269f8b46744cca4e584d98bc3fca6
a8078a4729c325b7504683662078036df59b63e1
927863 F20101220_AACRDS li_y_Page_127.jp2
a35e0b67029d5ae7aefd9154d045af55
fbca5e7472c50f777816ec7b6980ae410b539cdd
1046 F20101220_AACQXY li_y_Page_132.txt
d34dd812544c2c3862e33ce02b9391ca
324ac7ff6532e1ad2001643d344f3c6c6555003c
F20101220_AACREH li_y_Page_058.tif
3ca3cad291be2e97d03e02910a8e3c54
b11b07935d0cbfabcc802407a04f69bf207a39f8
43435 F20101220_AACQZB li_y_Page_137.jpg
b5eb6b00fa6052c7a4dc8d1ba76b4648
8edaded89905646a9dfe8ebcb50143f586880cef
2888 F20101220_AACRDT li_y_Page_151.txt
b61be7dadd8e95536dd464b55586eb78
ca5528967e197b38c420fc393b572a31c84b81b2
F20101220_AACQYN li_y_Page_022.tif
9c37af249e565ee0b7f3f1875e09d60d
98e3fdb0712bc2693032d45c11431d1dbfea6db3
F20101220_AACQXZ li_y_Page_001.tif
d6278e03a927bf62e87d78cd4be769ee
de6c8ca6ee274fd6706da17d1304e4c4589b8c90
1588 F20101220_AACREI li_y_Page_072.txt
5de913ec59c0755aea0f233fdd5363eb
cf56ed4ddb61e2b1c0c28eb13ae69a55004135ed
3655 F20101220_AACQZC li_y_Page_028thm.jpg
0e436d3c35357d83b5128b4ebc5bb679
fdd8c467e474f76c6e45b67d06dc9050bbe39227
F20101220_AACRDU li_y_Page_136.tif
7831ed359b63a176eca083d3d94d135f
22658bec0ac04df4208c3415680929a5ef80cba7
1051952 F20101220_AACQYO li_y_Page_050.jp2
1e7c41b7b824a7f6a16d85397c80a51f
7370dea99936fee1a67c6a5a04738e8a5fd1b96d
28177 F20101220_AACREJ li_y_Page_098.QC.jpg
94c4f20fc909e81ceffe068e3bdb3e3b
cfc0c26f23155b4b87b8ec403950e1eb27b8e3f7
1844 F20101220_AACQZD li_y_Page_103.txt
3140baf5ec09559128aec689fd723f23
eedb124752c3e4a71dee82fb21803913ac242117
1535 F20101220_AACRDV li_y_Page_046.txt
b8c0faef36da5d77aa839ea45b768783
0a161e9cfa5130c905c800528afc326ac39b3954
930 F20101220_AACQYP li_y_Page_079.txt
64293ca5797421de76195704823fe62e
60aeab4cfe32e2996d5438cf90aeee89a245395d
901437 F20101220_AACREK li_y_Page_083.jp2
03b9a44e083d3f16f2b109528e848cf3
4445f7958d1bf5818caff0c4819cf9af63d674cd
6488 F20101220_AACQZE li_y_Page_004thm.jpg
305934653fac732f41147bd830974316
474dd6f52fcaabf8501c63c864c0d821457bfb99
36039 F20101220_AACRDW li_y_Page_105.QC.jpg
5160aa5fdc2cafbe87c430c2927ac96a
2b52d3dae332eafc3cff875044bc7fb0995a0fd1
16319 F20101220_AACQYQ li_y_Page_090.QC.jpg
0eccac3f536620daa89a2e159618ff0c
820979be200bddf6d02821c371f046aa9dcba266
90641 F20101220_AACREL li_y_Page_081.jpg
9c4bc10b7f08796660cfb06bde16435e
329aac20a96515fa0a796b34afea6a11d65fab1c
8963 F20101220_AACQZF li_y_Page_018thm.jpg
261d061915d87bcb3fd8403b39734305
ba8dfc0a10a88de4109020616d951f1cfbe3f933
20581 F20101220_AACQYR li_y_Page_087.QC.jpg
bac47e317d683c381a3cfdbae3e9fa79
e238562ce97ba4bbee00b322b4f3f5ed02ea4dd0
108320 F20101220_AACRFA li_y_Page_069.jpg
4738df676fa8137cfaa68c3d099bde53
131d89850c566af45960e0a9a5f903a8da0b3571
F20101220_AACREM li_y_Page_072.tif
3e44b541691c494d791f46d37571c7c0
697dfdf9f576436372acb3b8cfd467f353741116
50542 F20101220_AACQZG li_y_Page_104.pro
5c339cd7f5a6518a5942b050ec3bf5e3
b92fc86e3862dea99eb60c2adc31a0de15df5eb2
1774 F20101220_AACRDX li_y_Page_119.txt
a8bbdab336c077302082c8698c52051f
9009d366fb8bc693dd120107356da0ad3b93f8ef
1182 F20101220_AACQYS li_y_Page_135.txt
b88453af591e43644002b66ba1916922
5285142a4487017548035e388d8a4ff97e7ee50a
40813 F20101220_AACRFB li_y_Page_083.pro
cc7c8f1adb02738fdea618d32760cec5
bbfce7c744684ebbb1d93b59b8dbe189a4735547
19280 F20101220_AACREN li_y_Page_056.QC.jpg
5c9e8e5f5dc20eb20b8064c1d9758cb5
3db57be93bc261ef3d998afbf2c1dc12c0bcced1
F20101220_AACQZH li_y_Page_107.tif
0b820a51ed6d8bb08f01672b6956bc89
02b49d870506b1427b3c85f7019c870fb9b13753
6449 F20101220_AACRDY li_y_Page_130thm.jpg
f37d129e9060ab5b13a9d64a2ec027bc
3d6e89022f6360ccd4e8011852beeb1138a04fa3
6173 F20101220_AACQYT li_y_Page_115.QC.jpg
2ca48aebc9e28cbbeac575580b7c94c7
34603ca143544141905f2eb46eadeddac5763eef
54019 F20101220_AACRFC li_y_Page_019.pro
d2a6712a9e7f5eccd80239a0a7b8c2d1
ca71c2f67bc8038160e89eedb27e9356bac9e9cd
8005 F20101220_AACREO li_y_Page_109thm.jpg
4f86689af555b3243eed7f52585ac7f9
8b1c98c83579f61130ef375b5ebfe8e2d5696e2d
1051941 F20101220_AACQZI li_y_Page_069.jp2
a3a5299f0dc7826594172bfdc9c9bfc7
d488d9f0d224c06d07fbd1dae1a41f02a0d34e12
1051932 F20101220_AACRDZ li_y_Page_143.jp2
d739f85832166d1b003838ea7e170630
7320cb42987609226b6a4767f579b1479d085d54
62244 F20101220_AACQYU li_y_Page_117.jpg
4883b4ab3bcfbd9d1cc1cee1e775bd08
8d0ff86a8b4e776d370a07cfed29efa4093cabbd
F20101220_AACRFD li_y_Page_113.tif
43becf9c67cb4c2ddab09bfa93f1df50
9d31e6b0d331e3537c8caa889b1feb268c0cb69e
F20101220_AACREP li_y_Page_004.tif
8a7f994ab7c1ec6c4837c7522d7f2077
697640c4ad9a507821564288788e71242519de69
F20101220_AACQZJ li_y_Page_067.tif
35783b22d8d11adb01be81d097b8e643
4b77cdf972c8d6bf89ccbd49231ea7110f4598a8
F20101220_AACQYV li_y_Page_032.tif
b98f285e490596ed41d70accb57fce03
f81713dd8ec138dfc756ee610ccecbe11ed5009f
15090 F20101220_AACRFE li_y_Page_114.QC.jpg
f047223f4b3986ba00f5a5262b605c2a
8dd7607a1c01415402e987ba3e20f75743bbcd9d
16907 F20101220_AACREQ li_y_Page_136.QC.jpg
7fcae774cfa92701cdc3b0792dd90253
874e8acaf37b78c0c66d3a36b5a7eb6e554347ec
F20101220_AACQZK li_y_Page_049.tif
c54c0125b196b678a9d5e7d87fddcfe8
325c106093eada94f1a09927161fd2cd2405f44c
936193 F20101220_AACQYW li_y_Page_130.jp2
65bfd2062f7cc0a06e8fdd4fadb4d6be
a0e550a22fc602c6882170ec32d9296fa146a569
542249 F20101220_AACRFF li_y_Page_078.jp2
5c830b185635b6b42864909bb0d27d6c
22627c9bee1a94cea4fc8d0348badef7a3cc40a1
1051974 F20101220_AACRER li_y_Page_040.jp2
b37beee20b8dcd3714f2845fa51d656d
5953a546b816fb4c6483929cc12def37c1ffde00
28818 F20101220_AACQZL li_y_Page_081.QC.jpg
e83e989a853cf4e0d563a9d5a982f405
9280766b03ef8d8abaf14feb20ac03443437fc88
44734 F20101220_AACQYX li_y_Page_119.pro
d489f7154deeb968035c315b05cd02ac
0a4cb7397f90a7fcfdb105deae2a28075fb9d269
2304 F20101220_AACRFG li_y_Page_140thm.jpg
d10800b0acde40872d5c61919c898e84
132aedc91746e844982d27c7fdca713b6f27ba47
F20101220_AACRES li_y_Page_114.tif
e898e83c65b45d29768527cc0fc54ca2
b147f65082f32bb06bca02f11315cadbe74d090d
F20101220_AACQZM li_y_Page_144.tif
a0041ea59fea6ed3cb16ee984493fb15
8ee20e310e361b30d8d71ef3bf3b08b09719a10b
24001 F20101220_AACQYY li_y_Page_116.pro
e61479344c037b43ea4a289c6c21dcab
fb7f6df429397fc685fc49c2b7418849a5c34638
3845 F20101220_AACRFH li_y_Page_041thm.jpg
f0ef5704eb0eb7db2e8126abd2ac6a40
42e20b8c15acbb72a7f1ccb67b870f0b7359f740
7631 F20101220_AACRET li_y_Page_085thm.jpg
34a1693397c6f74af596938f679c8897
09200fc2f7de7891bd68120deb7e8fb72685c93c
7064 F20101220_AACQZN li_y_Page_091thm.jpg
2470de19facaa4400c73d78175fd453e
192b04855a99adfe3de960feb9b721590c17e84b
F20101220_AACQYZ li_y_Page_085.tif
7224cd8b327c0bd6898fed2ec47babff
7cc90159906c20d9c89985f31f053a0001f336fb
15714 F20101220_AACRFI li_y_Page_123.QC.jpg
32ad9cb349e732d5a560a5d206dc8ab6
7eec493ae4bc3906e65e519f762b7fe7d5f73795
45630 F20101220_AACREU li_y_Page_077.jpg
0fdd8199898d552c99f30357a2d73d5f
bce806b55cf3accdd2bde5eb3f2e5678a1bdad1c
13884 F20101220_AACRFJ li_y_Page_115.pro
261fe2aef2326a53adf35fe9581bed98
0d9123ac6b00264b4b352851ef1072c1c9c72bfd
1051983 F20101220_AACREV li_y_Page_031.jp2
0264dca85d4314a379ad5d152e3cc717
ad3a2f1f3d97e59837ee6f66a2c2c22c73ad5aaf
879726 F20101220_AACQZO li_y_Page_030.jp2
f4ad56e01cc9f23c8cd51e77445c5103
bb71f975553333e2b7ddf300d2b5ff1b7690433e
5931 F20101220_AACRFK li_y_Page_119thm.jpg
937ef7a8f0ac058a818e933e1affe117
3136240b94a8a97b90a16fc89a2b4ba8310f38f9
46980 F20101220_AACREW li_y_Page_033.pro
c3881c7f9f8d69ef62321ee4f249b193
8dbd3f2289702da3a373394857ad4a1d285fdcc7
903272 F20101220_AACQZP li_y_Page_004.jp2
7271dc9aacf08c0bed608dba90b33200
2b7e9eb3811c13f47547b19b06b82193c10cda02
1966 F20101220_AACRFL li_y_Page_109.txt
e78c09a3ebcd5f17ddd7f830c61171ca
71d0bc241e551a3750de4fd892e6150e01f22ba5
3085 F20101220_AACREX li_y_Page_088thm.jpg
b25e7934d8b6f0fa88ef5b8aa1964312
2ae636d7cf8cba366ece7313b5a9cab33adf0cb5
2752 F20101220_AACQZQ li_y_Page_147.txt
47ad56b93d8ba664836097211a856756
255a708035e909db0609977ca8badae25894a591
31572 F20101220_AACRFM li_y_Page_014.pro
bbbcfbca2f18807ddf705bea337b6dd6
496f66265447ef6af0fbe147734b1835ae80f15e
F20101220_AACQZR li_y_Page_007.jp2
999267ad1f757dd5ea1843c74757f538
3a3242b11885782a2d655cef71877782b6bf7ec5
27272 F20101220_AACRGA li_y_Page_083.QC.jpg
d80af64545ec6360a92e26a6142a2c5d
cd31638575391ec30a0b8a8c8d9b67655ff25b0f
11261 F20101220_AACRFN li_y_Page_076.QC.jpg
ad1527f773d2f22b4ad0c8b1126118fb
b6b35fae947da3411193dd1882cc30394ec587a6
8227 F20101220_AACREY li_y_Page_045.pro
ff0e8bec4fbb30d55afd81a8b41a82d0
90d5c17e0bd916221b4f943a9265bc861efa5eb6
14195 F20101220_AACQZS li_y_Page_074.QC.jpg
8a56b3b00b6a70a94db06bac77e9403f
e3fb2767f802e8c8b52f46ff856930783c42069a
46900 F20101220_AACRGB li_y_Page_054.pro
5a37d9f8c02b150d22b792c86b3973af
5aa000b49296b499460da72953e585aa0ceebad9
F20101220_AACRFO li_y_Page_064.tif
337787270e3e7e6e18386324dc68d41d
ee04a5cb46738a1abd8528562f5dc0efaa7fa492
837 F20101220_AACREZ li_y_Page_090.txt
47f9bf4200a341fa8113b795ebcf7f1a
815863d2eea90c86d0bcce57cf429b3ecf547cc8
58289 F20101220_AACQZT li_y_Page_056.jpg
d6d89e2b9ace0aed87b53b0d757ea342
dc099ecd98c66e5d1371c8f7baf15a5030a65df8
2740 F20101220_AACRGC li_y_Page_009.txt
4fcdf6b4c02511c113fdc8c12aadd5f1
c0fa01ba40f1d011c0058f6549ca184c1ebbb447
7595 F20101220_AACRFP li_y_Page_013thm.jpg
c0947fdda44dde7c98b9fe9d0a465878
df67ffff73d794dbce87312556d12f85cf94f071
F20101220_AACQZU li_y_Page_030.tif
38d215d44e4e785777b1b5bc3e825349
71ad3b755690c0e415f964f83fa0116fc1a37a8e
130092 F20101220_AACRGD li_y_Page_150.jpg
202e45468abd306f0f0591e2350387db
ddea98591ff03e1ba55784367f6d9f538ecb5ead
F20101220_AACRFQ li_y_Page_024.tif
6ff201941c38b5f2de3342f0412f230d
867518c91751c39c631812a10cfd572d81786eef
33856 F20101220_AACQZV li_y_Page_097.QC.jpg
fc924fdd4d5e71abeb68c1ae61afb317
b21a0d14366636a7a4fb631ffbde5d029b4b05f5
1051554 F20101220_AACRGE li_y_Page_051.jp2
6f1e69aee5d14ba2d498b55187b2d169
fb1e82b42b799478407757ae6cb5566e77305588
F20101220_AACRFR li_y_Page_076.tif
e628f7fe190dcf96c1a2e2ed7e8febdc
47fbcf02f556ce8895abbf4de374411239ed1104
1051969 F20101220_AACQZW li_y_Page_068.jp2
53a5823b30589058c4306958c03f4348
9a3b6e52ef15b48cae606d07608ce82a1c93a3b1
74765 F20101220_AACRGF li_y_Page_011.pro
14581dd1b89369d00e0013e91c6fa7dc
526234f36919d0b55501e25a5953ff9aaa4240f4
55078 F20101220_AACRFS li_y_Page_045.jpg
34b091eca2bbc967a57b1ad62423d03b
8521598b1d285c1f51c2c4f4c67abffe7386a572
83005 F20101220_AACQZX li_y_Page_083.jpg
6d500e47cf74b6f39aa146f2ae390d0a
c3ea76b49f5a37ace57592a30b58492aae729b5c
8640 F20101220_AACRGG li_y_Page_019thm.jpg
040ed89a3ea97587d9f6f242397cf3f3
ccb02849a5c70295fac9c540bc3a2dd796f2e2ec
F20101220_AACRFT li_y_Page_015.jp2
428366df50555eea024a47440c347037
009e483095a91f493aaa5f83e886804aadea16f1
43275 F20101220_AACQZY li_y_Page_085.pro
a9cb9bc979aa128093f3db1b50d2d714
00df7ca67066ca87f96532ad506189a9131f4cbc
1512 F20101220_AACRGH li_y_Page_073.txt
45f294acb1eb89ab2c82954d12b72a65
aba363ca3c25e3f2f4dfb39b44c8cec8fc098694
48521 F20101220_AACRFU li_y_Page_107.pro
0c56e6494257bb69ae3d366516564826
b173481426424d6ea0912b94dcfca5a6c46680f3
2370 F20101220_AACQZZ li_y_Page_042.txt
8cc0d84cdc20422b3bffa22257326177
79d3cda79573c81b91488fe8d8bba1ab84fbd1a5
1051975 F20101220_AACRGI li_y_Page_035.jp2
00fa181795c1bb5b079154023ceb8626
f5cfa5389db1345e36edbcb9d11a4b5e04ff0b65
32945 F20101220_AACRFV li_y_Page_036.QC.jpg
68ad29a08b9be3459dedf4c7aa69139b
877a6c8a63e126261609fe4879cab54bb8419518
2727 F20101220_AACRGJ li_y_Page_010.txt
8dc3e3c25f2298d7e2b4dbc3489a7841
1a5b7df3db098ac1d41b166349cb5313a09c9db6
F20101220_AACRFW li_y_Page_046.tif
81f4a6583aec0a579364b320f1fe6e96
1f8b4e3122c1d47fb40194667efa0c9febb7f326
F20101220_AACRGK li_y_Page_038.tif
a3ebf9b07d0fbdd85eba399feb267c58
faadf4040751a1949fda4ed3e664b7f2dfcbda29
8695 F20101220_AACRFX li_y_Page_035thm.jpg
734e237581fab913be3fdc75d3ad9922
f9f45e5fb0b34cfed595f3c845425f9603d79f33
1051963 F20101220_AACRGL li_y_Page_080.jp2
447d6f89aa5423d84eb47be46e9d1cf8
dd7fd1b0b2f585534b740dcb4489eae095b2979c
27719 F20101220_AACRFY li_y_Page_049.jpg
090e3bd39c988369dc8d43686084184a
54965274cc6a24f6040fb39b29fe2fa31e5e1d28
1051972 F20101220_AACRHA li_y_Page_065.jp2
ad606643049beceeb57f3c54d499a7ac
44b1fa384d4105c3ddbab322bc6b49fe4a1b4836
95090 F20101220_AACRGM li_y_Page_110.jpg
858a175bc2de1bda5518f9cde0cb7287
434af06dd381f5c3ad3f6d902baa578731a7fa5f
2191 F20101220_AACRHB li_y_Page_066.txt
bf8b620c210f66499930de0b3e92e507
4b91c47876fa60a23798f42c9e97a5903182043f
37760 F20101220_AACRGN li_y_Page_018.QC.jpg
951c241326993d2b61a9948713f87dfa
a7938c1ead7cffe5fe1b771c1151ef6b56e1ec4a
25746 F20101220_AACRHC li_y_Page_030.QC.jpg
bab251baaf28cff8563bd6cb882e29c7
51569624d6eccc1845ef051b94dc0d257af37886
F20101220_AACRGO li_y_Page_015.tif
46142f8dc00d8577f6cda3bda6dcbba1
9b0ebccdadf2357a86fc801aabc28a0e88b50900
F20101220_AACRFZ li_y_Page_131.tif
a4ac79df93de2eb04b6d282cca01892a
d11914f66ac244f4c546133d99c3470d690270fa
37459 F20101220_AACRHD li_y_Page_145.QC.jpg
bf23009b38af3a5d77d6014329ea2b13
5aea19feee08ed13a501f602091c16c4bfd6f626
921 F20101220_AACRGP li_y_Page_038.txt
94d4c98b5c941ce32e11af7003db6efe
41b80a8da6dc7879f63e529b247a064fc801a995
17694 F20101220_AACRHE li_y_Page_129.QC.jpg
571b47efce698cbad97ee71bdfc02c88
52b7e563bdd5d9bebd5f7460ac3b706d0fa339ad
928466 F20101220_AACRGQ li_y_Page_053.jp2
903bd73f8a4eddb233faad154ce859a3
a27fa5408264fef134e31354a377a551ba19e969
5583 F20101220_AACRHF li_y_Page_028.pro
3cf57f8250222447ab9e9381c407059a
55475ca2174a1dbeab95f7f21c1e86c418b06717
37550 F20101220_AACRGR li_y_Page_148.QC.jpg
0e7c0ac4258bbedfb285fc0fe438023c
f7b08ec57d285aea002c8c4f7808ea4a15336d11
83285 F20101220_AACRHG li_y_Page_004.jpg
8a8f1bb57acd9310a427861715fb4bc6
60ef935cc8ec0ab09225af135afb4c4fff5ba21a
9209 F20101220_AACRGS li_y_Page_023thm.jpg
a1775ea26ba2b9b554c75fec1ff2f4e8
3739e61b73aaf7924b32b39e8d818aaadfd486de
97075 F20101220_AACRHH li_y_Page_051.jpg
d8a9bffd05e2a3e38301714b924d5aba
81bc40987e509ab144ba34791a3ad5e2f92effde
39088 F20101220_AACRGT li_y_Page_023.QC.jpg
1c431ef31c922ebf1bce9aaf73afda51
918c27a5004880c77a71d58f17d0034442ce087f
66038 F20101220_AACRHI li_y_Page_150.pro
f5dcaa006ea7edf98f56828eb00a584a
d7beae1c83f0af9da007e64ab01b26ab1a17e399
20919 F20101220_AACRGU li_y_Page_130.QC.jpg
1a2a82b386b01711bed195e83c41f7c5
5c3a0049a591ca4c334bf9c9fe8ce28179d12777
5295 F20101220_AACRHJ li_y_Page_123thm.jpg
8162f54dd4528360e3b340cb8a231cc9
e3d7597861fd0dbc4dcfdc8d3ebd2d68650ad300
1942 F20101220_AACRGV li_y_Page_013.txt
55424378489506bd60b3bef506988e09
52649916dace3001d8eba8888c261121b75021da
63671 F20101220_AACRHK li_y_Page_146.pro
256354f8eee1adedfb4103ccc796d91f
d93c2adb1484323575b7a458f9df960247c91176
221262 F20101220_AACRGW li_y_Page_001.jp2
47e80c6ab3574b5e1f719e441f857ba4
a1e572a346cb4c7348b697e61fd27c151049cc8d
574631 F20101220_AACRHL li_y_Page_057.jp2
c7840377f64b1f4a92d28733a23cdfce
515b5f8c3cb8fdf00a08470d79bce9ee6bbdc160
1847 F20101220_AACRGX li_y_Page_053.txt
fa51b38c2bd9bc00745a85ad5faea3af
f8670f6e8fd5366967a2e4902f613f5a26c3b69a
5928 F20101220_AACRHM li_y_Page_154.QC.jpg
2337d8e2ef9b19b3d3cbeb66f5c83911
6669323bf23ce4f06620b5fc5e46432aa5eb2273
265309 F20101220_AACRGY li_y_Page_155.jp2
8c8cab40475afcebc2c2a59be67e3b7a
a3f84a6f52502f1d57961c94c168d03f1155de30
20684 F20101220_AACRIA li_y_Page_115.jpg
68f699119205108acfd262869f3bcd37
668abe1d49d6328483a0c6d5c31d4e595c51a5af
1975 F20101220_AACRHN li_y_Page_032.txt
05a6b4d159437f0db4f0b2a1280c16db
eee78a7a85231c7121cba11004bff4d7282faad5
32423 F20101220_AACRGZ li_y_Page_031.QC.jpg
a78edd1871cdfd375e694f24a5f72d5d
ae0a02851870a8eda44be88e09276fcb52284984
6117 F20101220_AACRIB li_y_Page_047thm.jpg
b0c79c6bddb65df7832bc36bc8cb368d
58e654ac160feb553a0bcef0b90064b669c0d269
20387 F20101220_AACRHO li_y_Page_135.QC.jpg
e99c40464671f5c259d637feb2c38381
310b18c8e7c387bc88793cc826a47d8f08ab73c8
1051970 F20101220_AACRIC li_y_Page_019.jp2
0af3b5217be51e8e6ad54f71a510113d
3df0eebfb063651e8b49d259fb9c0d78bacb81d3
F20101220_AACRHP li_y_Page_081.tif
3af8b30950aadf74a3c553340b236a6c
6f5389185506b1ecc343d4e9cf8c610f06395773
F20101220_AACRID li_y_Page_131.txt
c231cfd06f271102a43fa61e2b920700
1f4cc26d49c346cd9d8908baf1e7e1478297ef46
F20101220_AACRHQ li_y_Page_106.tif
708ac5da1f3d504a6a219c0f4ea7afe6
d28033a65a419de4182753193234d4d38e79baa9
41652 F20101220_AACRIE li_y_Page_100.pro
33dd7f9a5db62c14959b91929433f012
acd4d602017395369fd72697effa23fb441accbf
49473 F20101220_AACRHR li_y_Page_036.pro
2338289ff1ac39d8fb916a58fe2638eb
1c6bbb1c3f6edc47a3c5ab52ff03e84083c787e6
2131 F20101220_AACRIF li_y_Page_022.txt
796092a04a870df71ad8bc5f348ed498
a5c5cec1778a09bd6b79b18a2f0b0ff3dbafa35d
1783 F20101220_AACRHS li_y_Page_094.txt
ef71c3f5e583319082b65208c98562ce
6d4579310cb31c5a0bdf9f7bde663559d08da92a
838524 F20101220_AACRIG li_y_Page_118.jp2
4490067418c04b1fce4568b413f2d00e
85ba1e7bb293ca0f1096329999b4078f928e830d
8281 F20101220_AACRHT li_y_Page_096thm.jpg
2adb1d9f7e742c241dbc08567669eee1
a6efea2dd949a9c475c6c8f030d44a22bc2cd6a8
F20101220_AACRIH li_y_Page_080.tif
56aae8355f1a0b8f951b04b742029733
b666edc395f6414abcaf416656b82c6a9705501c
6830 F20101220_AACRHU li_y_Page_030thm.jpg
8e2b9e7691a13e61663cfb4b21d3b511
4aa091185ebfdc4d090637a0de24085f5b484f35
1051955 F20101220_AACRII li_y_Page_036.jp2
3cf80b072b237e685deddda304c73998
76fe664fcdc9c52c5d52d0ce503984978ad3cc39
27166 F20101220_AACRHV li_y_Page_044.pro
15179c042131aebda2b2b7a546685dbd
00385ed1d733d75ac77ea691531108190247023c
28710 F20101220_AACRIJ li_y_Page_061.pro
8ac5bad6f69a884d8cc40e78f8b400fa
fb94e716db218318e1c529a6b7b6e06e29d4e08c
F20101220_AACRHW li_y_Page_039.tif
ec6d97d7b2c07a5022f90db58dc1cdd9
520b80552a79354c6ac5dc3c13ca3e03ec457f48
162736 F20101220_AACRIK li_y_Page_011.jpg
638016f327a8f822e103402fd64449d4
876b249e4700136f7b599a04286b22844f5ead47
F20101220_AACRHX li_y_Page_041.tif
81d695124f04193d3fe86941d8e37896
62cfb17e97e9ace9566441cf000293807dd89226
F20101220_AACRIL li_y_Page_087.tif
2f318b41d93eab9273de11a06bc365ce
fba65d1fbf8a751e429a745548a8fb6d39e5e0dc
F20101220_AACRHY li_y_Page_074.tif
4b66f612e28634c2d786945f5a0041c9
324b6c589fa4b9e40cf5157928a62ee6834fe157
52017 F20101220_AACRJA li_y_Page_138.pro
3b8ed8de19460d5217f0704aaa0302b6
24af0c2bdcae3e584c15f796cd5d6201f0b4ba97
F20101220_AACRIM li_y_Page_017.tif
fb3d0fdfea11a40e782ad6946504a133
f5936b9487c0b54b1f94262f9b6b2885b02b652d
1985 F20101220_AACRHZ li_y_Page_036.txt
4d2e7a2dbe76131656e122c20a7c28da
28f4904b45a7f5431051c4c0549cca1f01cdbee2
F20101220_AACRJB li_y_Page_043.tif
651295f26a5ccef3ccb814727695d624
ffaf69b73dc2a341a1cd68e6c8698279d3754f10
46683 F20101220_AACRIN li_y_Page_051.pro
e2128075d0f1f0678a906491d57b689f
5d3c0cc2a47f3022100d809b4966aed26ec91d33
2943 F20101220_AACRJC li_y_Page_073thm.jpg
1aeb5c477ac51a42bf16a143f9dfc6f2
82c07b467d2a6b400b9bf3541861b4728956d1a2
37128 F20101220_AACRIO li_y_Page_143.QC.jpg
dbdcf13f9ca3e7a516060052f3ac9c70
0900549244474a96690cdc54cf3e55437a2decc7
28055 F20101220_AACRJD li_y_Page_100.QC.jpg
7de1c32eb5f4c655f9902a560556a4d3
168ce7d32415b12ec431dc8ff0078f8f1601d942
28193 F20101220_AACRIP li_y_Page_013.QC.jpg
a96f97d60ac181a663353dd89ac3a3b9
c416e0943dc797c471921834a2f78833cb7dcce4
1914 F20101220_AACRJE li_y_Page_107.txt
3fe457e75ecd60893502d7f9cbeec0ce
0055db03e3b394822b5439fa5813e80b5dbdd74b
9011 F20101220_AACRIQ li_y_Page_133.pro
3e02dbafdaac00c850dc15ac91b2dca5
b76e5619f4f403633dcdc7869225f804499af725
2171 F20101220_AACRJF li_y_Page_062.txt
a5cf98c035d03729daa27a05ddddafb3
4666f1dcc7761dba9820a0aa13c4e318ae553d3b
F20101220_AACRIR li_y_Page_088.tif
3ea95740f4b6c1e6f584d62ee84c618f
3c4ef4545a2932a2098c5850047a356aa94107d8
F20101220_AACRJG li_y_Page_138.tif
8a4875f4ef1d1a807dc4ea92eb74d30d
4f6cccdc3d3359215d1609401ffca2477b9ceba6
29911 F20101220_AACRIS li_y_Page_086.QC.jpg
a364e693506829a663a184fa4b3e88be
64d397c0b66800a21fbaef58889aabda61e94ba3
1583 F20101220_AACRJH li_y_Page_074.txt
4948f4fe63fd0f4f5e845eaefedb6cb4
f16de5b458a3b606dd115b8ff11d6ffd9c84b505
17957 F20101220_AACRIT li_y_Page_012.QC.jpg
ac880494ad2346d74f6bf77b48caa779
48d37df55b5306bc631ac13317a9af514c3c56a0
F20101220_AACRJI li_y_Page_121.tif
a6f3a4f9ea75c4973ed9dbf00ce6a88e
5eaf1643572d60f805fbbc81ab22aaf94487c4cc
8653 F20101220_AACRIU li_y_Page_021thm.jpg
ebae89d46f7cffe9abf07ecee307b8b8
53306d60a571d74cca394f54be5c0f0e91eee7ad
2848 F20101220_AACRJJ li_y_Page_144.txt
13e0a8ce2caf6d8bcc92d60eff1b6cc0
d6547b00216e9a219dd361e0776e3c74a94a2e46
7123 F20101220_AACRIV li_y_Page_134thm.jpg
85be6cdafa23d7489cff1980b8ecc5e3
4d8017091a8cdf8d23e273de7a80853841294dbe
F20101220_AACRJK li_y_Page_145.tif
df940b9a5b03ae39f6b8da7cb944db7a
987396d03d1311b5d37e75acb9e1f324985e7a31
36960 F20101220_AACRIW li_y_Page_150.QC.jpg
044e66a41aae16a552ad9416fdc46147
9a14b5c06c873896d06cc70e7153478db3719809
11015 F20101220_AACRJL li_y_Page_155.pro
cf1540ddd4cf34c5fcf38a2ca01a6db9
c993997c38546cd1c7932452e2882545dc00389c
1051961 F20101220_AACRIX li_y_Page_023.jp2
cb5303422116ecca998f555db951aa80
6549e336a3868c386d2702456ca14a276164a428
124791 F20101220_AACRKA li_y_Page_008.jpg
2d6050927aeb21132116dbb7ef1f55fd
d646bec4305c86a4c51780fe4e7accf00c345556
13542 F20101220_AACRJM li_y_Page_126.pro
1be18c13a6c8c2a561370b09bba2b57b
caed564144fe7bca4259d50149bd4231f0b2b8af
2681 F20101220_AACRIY li_y_Page_148.txt
395166f1b2bad2bd4dd345c12cf2cdd2
624a180850f63d9e588d84701897e9d1db21b7b9
64712 F20101220_AACRKB li_y_Page_149.pro
b8138509ce0a4659c2e200bb5ba77d11
4edc37f509f5279c0825b517c2b8ac7e8127b8c6
9439 F20101220_AACRJN li_y_Page_144thm.jpg
c48f3cd4b38dc2e777ddf61217d9dc8d
72567e77b5f91a16a926d02c11d8d4e5e131c38a
33416 F20101220_AACRIZ li_y_Page_016.QC.jpg
490f083ea24d914577d4a0b5194851ac
c7b6ab26ef9c00c907aa0914baa6397b328a3031
F20101220_AACRKC li_y_Page_154.tif
813f133b6705ee5ca3b9c87511890969
42cbfcb98d68b58ebad7d0bda870fd99277dc3c5
1051909 F20101220_AACRJO li_y_Page_142.jp2
cb0927cd5224d7f5be7d1fe3b1a6dcdd
c0449d5e3ca2c2f3b3b771cd0ba33ad0b1e1d4d9
107689 F20101220_AACRKD li_y_Page_017.jpg
aecc9bf15ec32b94379b7be9f720bc89
7c5383e8c1c352ca11fdd4a754b4cf714af8c526
990707 F20101220_AACRJP li_y_Page_086.jp2
1da1a453f0676266d8580b065e939a17
bef21d7111b1864b464c48aa4ed1151ad8f23e35
F20101220_AACRKE li_y_Page_103.tif
2fb5c53fe955be07881d63e8f593dcf5
72d507db3d1330dea8932ac896b3fd9e899e6a6c
43955 F20101220_AACRJQ li_y_Page_120.jpg
1b34039606c643c21c13a9d2a4793bd9
298bed2e403c1f041f7fa33e9194e79307d2f618
41148 F20101220_AACRKF li_y_Page_122.jpg
b6e39fe948408179b67dc9ec0d8aa2f1
086fe873c67f05c76a8c4191e9e8847ee74b92e6
49348 F20101220_AACRJR li_y_Page_015.pro
8819f1bf25ac788c625949276d34f585
42ea09f7657199b1df2815074cbd64db7972f18d
134328 F20101220_AACRKG li_y_Page_148.jpg
abd490fbfc51d1520f0eaa3f5e8e6d69
f648ee63e62abf8dad4666c04a77bf319062e2e5
1913 F20101220_AACRJS li_y_Page_051.txt
f24ca8363f72ced8ced1e54d4e53ea4e
440069112d39149804ce8fa9220a1f4446a2fb06
25338 F20101220_AACRKH li_y_Page_118.QC.jpg
4ef90a4a90d9dd725064e247a54fe18b
595773c084d6d3b0eb7f9a2d9fa744c9c83df893
46992 F20101220_AACRJT li_y_Page_039.pro
cfec4a2337f1df889296e971b66979ec
cbc9a5bc8cd1c521e7289f536ed6a9748e2bee04
20638 F20101220_AACRKI li_y_Page_078.pro
f5fe54f3f917240c2d3aa8f242c51b60
a9d5b81b63b9a4300d7840af74637e9960fbd3fa
54140 F20101220_AACRJU li_y_Page_043.jpg
1dbdd9f23188abf9750c3f92900127f7
1531d8db75ec661ad40e4e95b59916c772c7e3f3
1836 F20101220_AACRKJ li_y_Page_101.txt
f7742f1566ca201819a3496ff0f67591
0fa0a7cfdb54180d0a98bd34b7f97205b4a36272
5999 F20101220_AACRJV li_y_Page_118thm.jpg
25aa2515cf2f6321b70290ca3d10b4ad
66f250b82b0354eb04290e0f351bd601ac2603db
18561 F20101220_AACRKK li_y_Page_078.QC.jpg
60406dd89dd9f1f887500d45a271529a
1ed07bd10450039433940fcbbab05ede4fae4be3
6324 F20101220_AACRJW li_y_Page_079thm.jpg
e2200aa373802994cdd9b1f7ad3591b2
44ecd36dfba88bb1067aefd2863e6df5543cc23f
F20101220_AACRKL li_y_Page_142.tif
738a187033bc72e6a5b68ccf4f0e84d2
63e14ed7ac7f9f8c17169cbca376e730ca3f4adb
756644 F20101220_AACRJX li_y_Page_063.jp2
cfa25917f70094f3e59d3b6a95c535ee
c5fc0f62f691d6fb8cdf2b1d5b2d2dfbe4b82fef
2678 F20101220_AACRKM li_y_Page_149.txt
efaf38be07d22088c07c531374e24078
66f39d761b98571efd2bf07878e5c28675f73dc0
686977 F20101220_AACRJY li_y_Page_046.jp2
4b5da0795c557b4ef2c6bc48bc410d79
064a912b065caeef72b3eedec282e984ea7662bc
23499 F20101220_AACRLA li_y_Page_001.jpg
d175a3fd18d1119fdf6b72555bd2262b
55b9c4f7a82f984a6848a96596a5950e0e71b423
8931 F20101220_AACRKN li_y_Page_020thm.jpg
c93e457f701cf04cd95cb6d0b58f12f0
01ccf916a293b976a62fd67d35bb850080fe6387
7783 F20101220_AACRJZ li_y_Page_007thm.jpg
25c0839b2d5bf25bead5962c59fb1064
7ea1f75d6a0496df3a194e32775e34549848cb64
5825 F20101220_AACRLB li_y_Page_059thm.jpg
2082481f1c164eae08314e9bb1cf7974
8c73745612f735410fc9dcf5ca79054512914efb
F20101220_AACRKO li_y_Page_023.tif
8cb3ead7983f7afa2c931c6a9161a748
97203dd52eae4cf3ee372d441b550575f62369ac
F20101220_AACRLC li_y_Page_014.tif
05b26bfbd21f8afbea7968450853f573
ed8c76f61ac22e276cdb51f5a7729fa76a14e715
8439 F20101220_AACRKP li_y_Page_097thm.jpg
54b66797458ab7284d9e98c68e72a844
d5e16f6e3e44b428328ad07a316e901afc923529
422674 F20101220_AACRLD li_y_Page_077.jp2
0108b504191b2ffcd491af3fce252de1
643478fffbb94e4515e0f5a8b4c65d3c902e3d36
43450 F20101220_AACRKQ li_y_Page_081.pro
92ef4e96c77e51d9f5962c3c396b5970
b73c5ed71fa3700dcc2a62ebd4c250f2fc9470c3
1051973 F20101220_AACRLE li_y_Page_096.jp2
1b3a0ef00b4973d9e23387632dfe01f2
125791dd65abffed765baee77d1a25a2cbd3faa6
50496 F20101220_AACRKR li_y_Page_096.pro
d1df90d2af501376f3d12c116e3746e6
0e330d1491afd4a5cad9d41ab0ef3f474bc9eb04
64523 F20101220_AACRLF li_y_Page_010.pro
c9c7fcf3e688e17c410b3162089ab9f8
ef64b8a8ac8b23fdb03cb36cce66ac566aa380f6
477789 F20101220_AACRKS li_y_Page_136.jp2
32fa2578292d3adf62d713b35ffecce0
48703c91527eadc70e8497625b56734ed97e071e
F20101220_AACRLG li_y_Page_100.tif
23fca546a4f920bfb6c89416dfe32295
53326951b9a22f67f4f1e39540659b5a679eb4c2
70912 F20101220_AACRKT li_y_Page_039.jpg
f57c16353bd277e5721aa3902a231b00
fb40d44ae388f6e6796a63aeafd4e078b5d7e4e1
288202 F20101220_AACRLH li_y_Page_140.jp2
57f606503123dbfd3940786066314e53
44091b241f06d3ee926af9ccac4bc80b504b9b39
1866 F20101220_AACRKU li_y_Page_003.pro
0fb21f096135c2d3d3f94e98de563bed
2f21c26c5f231982d7d7c7b9f7092b70a80d673c
F20101220_AACRLI li_y_Page_086.tif
007e827e6782a30c93c666dc3a8d42de
b97d691f5348eeb851016094ad5d4a84dece0d7d
14901 F20101220_AACRKV li_y_Page_134.pro
bc8c7fa2b0e5798c95977c69c13ca275
250d800f77555490d0202745c0c15d184907b267
F20101220_AACRLJ li_y_Page_116.tif
14613bd74eae5811f5af3f77440f5607
6d95a5226d9c2ede3c65ed6f1e9a82f523499156
1051922 F20101220_AACRKW li_y_Page_152.jp2
19eb9edad00b2cb66b2b0d96b8a241ec
aafd23400c2408986074281407868abe95597cdc
3098 F20101220_AACRLK li_y_Page_011.txt
7c45d97456aadfd15591090a06d47695
8fe6a168acadb6d95c5dc16648d8d53d22271e94
1051971 F20101220_AACRLL li_y_Page_010.jp2
5d5931bf06d0140d08169b8b527a35a2
f29b90b218108a394939d18c8907d9204fef03f1
F20101220_AACRKX li_y_Page_055.tif
38be708c84cfe91453fcac94c851069d
734eacd7db6b85c626bab2ef3efff1398ca06f19
81702 F20101220_AACRMA li_y_Page_037.jpg
3647869fde22b248b5ff3f90ee7c9b59
c2bd648ec5e254019b07a7ed8293bd83aa3e3e22
F20101220_AACRLM li_y_Page_140.tif
acf54a826cea0d7c40ee9184a4242171
d6b5cb7df3157d049e6471870a4bb65918f5c8b7
896483 F20101220_AACRKY li_y_Page_073.jp2
8e95634f001a3b8813dbf2932281c0c5
1da539fc05b2620b52782ce181c11c800895764b
28175 F20101220_AACRMB li_y_Page_099.pro
b5e760ce89a1d983e7ff82c38f2543b9
3dddd31521e3e17e19fa374c9b5114420ddf72a1
21645 F20101220_AACRLN li_y_Page_027.pro
c3da71d250adfddbbbb931b5e21f6d5e
bf79a427adb05530911d2834bda87a238063351e
48863 F20101220_AACRKZ li_y_Page_097.pro
3bf271093cf39434916b27710e595cf7
91231f1c9c8e6201ed47af20c625905ddb25395d
152398 F20101220_AACRMC li_y_Page_007.jpg
41c9b54e8d7b37ce30878e326fd1d800
d544f6d52ff5ec93b7c49a554fc2afc0b199dc6f
709 F20101220_AACRLO li_y_Page_130.txt
708cf60fec3e4b09dd66856383ecb984
2133c589138b807537853da72559137866246f61
F20101220_AACRMD li_y_Page_109.tif
86cf887b384bb6b2f81f1916754542df
d480d8158f52df26ffb5f30a3c9dae4e8067d36d
842910 F20101220_AACRLP li_y_Page_043.jp2
f6f0f7f02e066ac6456e090174be389d
4e509e9f6a97c558fb3409360610d3e4f5c91d50
2032 F20101220_AACRME li_y_Page_035.txt
b2ff9fef4fab229e4c689574bfe9bb69
2ae0907e8041ae932d9ed41fd8bb0af2a869058c
83658 F20101220_AACRLQ li_y_Page_119.jpg
1f4d947d1faf17892b4359043d32995a
e57b81ef935cf33d6a183884df688ccbf8f64b54
8565 F20101220_AACRMF li_y_Page_139thm.jpg
d60a2c26576bd6f24615774c26df58eb
06486349e5cdb301c0e6443a9838e6a735c0ec3b
66868 F20101220_AACRLR li_y_Page_147.pro
bfc225068184e4e21f2d3d786bdaf598
d30d1d23db82413cbb1d0e02c1222500b2dd6e6b
24854 F20101220_AACRMG li_y_Page_060.pro
074776b7766395a85c63cf0981477221
6093427702e1d76b8aa8d7c7f4374d59cfc67df7
2029 F20101220_AACRLS li_y_Page_055.txt
5da657c4bb3ac457d327f7e2e2164386
28512a2d84baf3c94f9a2ad44236af0f7d4909bd
1810 F20101220_AACRMH li_y_Page_091.txt
058a1e1de6876aa60a4325c545eaf798
3e9e48d1177e4fe638f4346bb6aa4b56c824de94
101018 F20101220_AACRLT li_y_Page_031.jpg
f5f7982dc30a1d214b2099b5554f6c67
027d42b92779362f9c0060b60ff4b6cf9f01000d
F20101220_AACRMI li_y_Page_135.tif
40681514f3709141f25c8907b9f7e314
b40bafa8b9128d68b0e9fbeac71e039b00f41342
F20101220_AACRLU li_y_Page_152.tif
deba24f8989652cd367220908f4b9af4
b395aa1d00dab88c49d835aab8995e10664904c1
F20101220_AACRMJ li_y_Page_027.tif
2e38dab626e85112d21578ff33aa2c01
6fa0212c43f56f09f3ccf97760fb42947795577f
5023 F20101220_AACRLV li_y_Page_026thm.jpg
9793553442708485a11ea8cc0bd07489
bc0230480a368a2f986976e9713dbf4a00868c49
F20101220_AACRMK li_y_Page_010.tif
a7c31486461228efb670e5f5ee6492ac
eb1d93657332bf3ba125df9c2afc90128175fef2
F20101220_AACRLW li_y_Page_028.tif
d12d541b990771b65fea6a7459e77e55
816b08548f681959ef10b3d767989f7c724e3750
1007955 F20101220_AACRML li_y_Page_106.jp2
63c196257588129ba16c6f1bcb17e194
3216253a0460ea4697e1980c7685fa6eb0f26e57
3101 F20101220_AACRLX li_y_Page_121thm.jpg
caeb54398e7f17ae75effbe7d543882b
673cdf658d55714f8a97ed4d54daa5a7ab52233a
1051951 F20101220_AACRNA li_y_Page_144.jp2
8cc03ea913d0b15b018d1566282c476e
92ea6c16231f8162b10c0a167d8513cda055c29c
32273 F20101220_AACRMM li_y_Page_051.QC.jpg
ff4e5f66e7b9c56833b953470be11bc9
1b89a8e1fd55fc52ab1c7847e4da691f157e0da0
54607 F20101220_AACRLY li_y_Page_066.pro
ad0a17ff59cb4f3e604c43b310666a19
11d18099a62180d5dc75675a478499724fecb58f
31528 F20101220_AACRMN li_y_Page_034.QC.jpg
310aaf1c1681ec362aad07c4f1219241
5372febafc423995514657e4c37360f5f8b67677
597710 F20101220_AACRLZ li_y_Page_133.jp2
9499f2b31efac8f4db02549ab974e577
9fb21828d2ebbb54e65ebb5ba092b5870f98f7dd
37744 F20101220_AACRNB li_y_Page_153.QC.jpg
608b80e165f3e63cf7c442bd5e94ea4f
811e0d577e3d24f9be2dddc7b91302f19a12cf8c
45547 F20101220_AACRMO li_y_Page_118.pro
415575b6929d8e564be29cdf5271baf6
7df8d58a515ec727175ebe6b5d2f568dab6bb142
103650 F20101220_AACRNC li_y_Page_097.jpg
0fc0f4759b4676907d6038f5077f9035
2c197ce14a2a3dd89756a289c6e5a7dccf8461ba
35576 F20101220_AACRMP li_y_Page_149.QC.jpg
879dfb6a6c1068daef29c7f0ca004bee
f08fdd1d38bb17583b42e3eb1112e7398a09eedc
444983 F20101220_AACRND li_y_Page_064.jp2
e23be382f11183abceaef84d342dd397
64b940716a046c67079c31ccb02c624883415bad
300 F20101220_AACRMQ li_y_Page_112.txt
acac9584a657c002aad2ae74b285da2f
60b99964f40cde1d419e740c807cf8670b036206
5824 F20101220_AACRNE li_y_Page_126thm.jpg
58b25a69a11686b48f5d263ebb0aef08
b12966621d383eb14de454b5c11cfa8d0409a2cf
98158 F20101220_AACRMR li_y_Page_093.jpg
6572a4e3ce24048bb6758b2566ba34e3
e279f1f01cea4120867ce3f9f514392b830392ad
2199 F20101220_AACRNF li_y_Page_018.txt
89daf2de54da0dd45688845e791c7fc5
1473ce2eb5f37b34c6c4f343da23bd1bc84c903e
29056 F20101220_AACRMS li_y_Page_091.QC.jpg
7182e504837ed1d9830f9990dd7b4169
c07d1f93a6b8ee4634813dabb49236da5a152c02
1087 F20101220_AACRNG li_y_Page_116.txt
eab349beb1f214c0f7afec2437176f0c
93034dea1faf3aed233ebffe5a5829991fce198d
5240529 F20101220_AACRMT li_y.pdf
796b42a7e8b162163fe40f80d4d3e999
4946adbf7ac606176a3c048f8b442b930b7b0fbe
F20101220_AACRNH li_y_Page_048.tif
836402151f323e67bf9dd2b41e72e355
a0434639cc71f84da31fa0a0de4e418a623b1e27
33875 F20101220_AACRMU li_y_Page_067.QC.jpg
5547ac17f21851d7fca50f084a2cc63c
0bdb7766ed6e5186b18c0a9d1643e3577e959862
3961 F20101220_AACRNI li_y_Page_089thm.jpg
9621cb8e2cd15d622a845413b624e2cb
b69c8dd18755f82dcf8cd04994d821ebe6dcdf05
8612 F20101220_AACRMV li_y_Page_069thm.jpg
5ba548a78c0a1567f75e15e757447459
e1ef2aeca3464a65d53e5f8158f15fb303f3a1e4
246313 F20101220_AACRNJ li_y_Page_049.jp2
6c2a27e67cdfffaffff6e659105dd749
9ef1301682b411018ad94ec0b02fd32328436b73
86357 F20101220_AACRMW li_y_Page_006.pro
df5125aab94c0efc90fbeea28b79ab06
7c4914bb8dbeb35223b167dd8d1c881cae7cf7a6
34328 F20101220_AACRNK li_y_Page_015.QC.jpg
a81d9e863342955eb6221722408426b1
dffcf573a8db4e25c0e3713347d5fa6dcbb6bc77
1761 F20101220_AACRMX li_y_Page_040.txt
955494fbdd88905966c4812fc3d3b8c0
39fd50783212708824aa9b56080756da1525d0c1
F20101220_AACROA li_y_Page_061.tif
7d3eec9e4dd945a89cfe33c8eb93a489
1c40d11265498505e56420283a73e22e49c76b9c
3294 F20101220_AACRNL li_y_Page_005.txt
cf8a4bff2142b695ae225dfcaa2d0acc
ac60b44f382d0a9c4a6b7b7747b27a85aeec8fa5
8988 F20101220_AACRMY li_y_Page_153thm.jpg
c2a3ced91f558f8e14849614ee082f43
1853ae84ab8da63526293317427d0811e862e571
8517 F20101220_AACROB li_y_Page_066thm.jpg
1a69106a69999b9d0b9e152ea9efee35
586e129464e5ec7e70b1706ba75f5fe3a116cfb9
35282 F20101220_AACRNM li_y_Page_062.QC.jpg
6db52a5f365cfe97a07514d965406286
da2047088244da870ef72dacabf3dbf3f988f3da
8345 F20101220_AACRMZ li_y_Page_032thm.jpg
2cdb3deb08288bb90ef01fdf7027c098
0e23e735f1cac7ef7dc1b962022f30dbc906f53b
1038066 F20101220_AACRNN li_y_Page_135.jp2
6d7ca991cc33624bd952902bffd1ddd9
ca4f376cb2c0617ba5995f71ec5bd384783c2eb8
1726 F20101220_AACQLA li_y_Page_083.txt
af47d1636394cc7e2ce3cf18555dd727
33aa4c9fcb309ff41b943cb47db3559c95af47fb
81 F20101220_AACROC li_y_Page_002.txt
564df9a043ebc566a2d883915b5143db
10191824e92311c4510c6bfe2bd41c4b60cbc163
106585 F20101220_AACRNO li_y_Page_138.jpg
7bc91e4f0e27d28a49f582a03d1627f8
e5110871680fc8ede7a316e7911100ce875e781e
61199 F20101220_AACQLB li_y_Page_127.jpg
fe3fe035ab0ffd0f2897683a5241f2bf
74f997068cf906c9cb430c79374e020236ec0b5e
1264 F20101220_AACROD li_y_Page_014.txt
7bb8262fc41c3dfa08c24fef0a56e95a
1a5879835361f277a61eb4dd09e00422e5d62343
6033 F20101220_AACRNP li_y_Page_112.QC.jpg
997b56f0210150bd555c2c07eb7cbcc4
764c5ff5cf38d7a574afc7bf9c153eaf5a1343ba
35059 F20101220_AACQLC li_y_Page_066.QC.jpg
c4b28c7ebb35604a7f06ad1d2cd54ca5
631dc0c8866a1f578fda402a20dfca33f484699e
1051965 F20101220_AACROE li_y_Page_006.jp2
54b9e6cb0fd163cdcbc00ad5aa3b37fe
a5b1d9f02ea383426f271d18092beb0d371d56ac
921248 F20101220_AACRNQ li_y_Page_082.jp2
b8e44b016eaf3b7d4289533b415e7054
e63f10728e0109b3c5b97d9a8ae1c9753c92cc14
2720 F20101220_AACQLD li_y_Page_143.txt
74b1649ecb035f615eccf942e8486e05
4c483467276cf12e9c5e92b78f70b7843d1321d8
4202 F20101220_AACROF li_y_Page_039thm.jpg
ce1a6067c128de14b3d43e3f99c1d648
822a9703cd0f07066ab6a3c730b5dd3ba66fc5a7
7823 F20101220_AACRNR li_y_Page_122.pro
f6adbf65b34c84864fc87871b0082206
be2b165c2c05bf166926f51432e74589082ce2c4
48404 F20101220_AACQLE li_y_Page_089.jpg
2144d8732b33f34c7547390cec6d47b0
dd43366d8b8d92f6069f161372e3dcd20a04aa9e
55853 F20101220_AACROG li_y_Page_018.pro
e49682cd54d7ee460139181901660770
5dd7a942ad380528b53f890e8b4f566ca45be86a
21775 F20101220_AACRNS li_y_Page_014.QC.jpg
d5e15a464109c5af33129da603bf9b1b
0ad15d75c6db1bb0bf1fed8305153a8fc7030f49
F20101220_AACQLF li_y_Page_146.jp2
b0eeb51f7c5856446087e96ed3eb957b
07bef8abbebce748e58265f3cbd7954e3cdb58dc
105456 F20101220_AACROH li_y_Page_105.jpg
56ae92e7d5cfe79fca6c990e76f4b4aa
f5d3315e5c5dbfea624d6678a23ee38972318c44
1237 F20101220_AACRNT li_y_Page_012.txt
8362302e1d58f881614836d2979aa5cf
4584ede4d1aef48d80091cf0a1b1758b2bf27926
660129 F20101220_AACQLG li_y_Page_117.jp2
852f3adfed54538e81d98418a7f633da
690241ca233a4bcb3069cb101ee7cb2f215f4738
F20101220_AACROI li_y_Page_016.tif
f7643c70481016c55585b85d3db89dfb
434d3f3b41023b91c32ea7dc765255494786002f
899733 F20101220_AACRNU li_y_Page_074.jp2
0fbffbb1d0c10b3ee6981f40978bbc59
1f6501ac2351a3d0a2a6fd628718ee0c073939f4
7369 F20101220_AACQLH li_y_Page_083thm.jpg
5f3e9b53ffdb6bc8c31c8189a9d279c7
4018497c34782d281de960bc98d80fd9a5b8d53c
28562 F20101220_AACROJ li_y_Page_117.pro
bd1383b3ebca070102ffe77c0878c4ca
be19d8a8ce7a1e5af8d2af2711f1b59bcda75c2b
13632 F20101220_AACRNV li_y_Page_128.pro
68eff68a4869f442bb625469275e237f
cdc6b0f1d53694990d19f4ce69b10048f9aede5d
15841 F20101220_AACQLI li_y_Page_059.pro
35ee95bcec63f8f6a348e9579504d889
ab82a398a956f1500fb569351a2cf0bc86d500b1
F20101220_AACROK li_y_Page_115.tif
257adcf30fe90b933cb4cf39799b46d0
ab67107ff98e86a6e688748dc0a20b951b40e1c7
10075 F20101220_AACQKU li_y_Page_029.pro
5faa9e33dd1d2a6643fbcc54064ec4c8
19676726550a7a487f71548cc9a84c2ad90620aa
F20101220_AACRNW li_y_Page_112.tif
858182caf7c62ff98e2c98509d8ac3d7
b7fcb389482ada3f84c75dda2fe262d84eb288b3
63998 F20101220_AACRPA li_y_Page_044.jpg
0f4d8d1c3a164f9eee463d7bf8a16479
e4c051b0e9690d5617db177508a524d96e10cbf8
51624 F20101220_AACQLJ li_y_Page_035.pro
c1bd68e0539f8649bcacdcca2794e088
26402295b953f69c35d8f8daa1b526acddd293f6
5287 F20101220_AACROL li_y_Page_014thm.jpg
f78e1c6321973ffcd09874b25014dfbb
648c5284385b24d7b8ef08a517e656ef9c6fca60
19285 F20101220_AACQKV li_y_Page_117.QC.jpg
87351c3a4b3e1964ae17c7ee6725246c
88b3c2115556de2139aadaea1261f6c1c351b95a
34476 F20101220_AACRNX li_y_Page_104.QC.jpg
6eeb1c2a4d401f56edc578d1b1ef12db
f49264b3c8a8f1a103e216cede92c5aeb3111bd8
44751 F20101220_AACRPB li_y_Page_123.jpg
bca3e2231098fd31de92742658eac2ff
cebb4f9e208e96b95c04a4474c2f025a5afb731a
1412 F20101220_AACQLK li_y_Page_114.txt
d54f2edef2de1a149589ccdace47a70b
e24c44ff6aad282340436f81161ad2bbc89e5a08
9437 F20101220_AACROM li_y_Page_151thm.jpg
61d6fad4bf1d2208f0b5bd07e5b79991
2c4a1e5cf1abab69548016467aea8dd1e53ff85f
34283 F20101220_AACQKW li_y_Page_096.QC.jpg
d989d778e0f5003cf1d3672c7f99627e
75bf402e9d8a9a4a609887bad510395df285c113
11404 F20101220_AACRNY li_y_Page_077.pro
d92975899796dfd9a089c1fca58161c8
b1250407f5604ea49f80031e4bf99411e1f7322a
4114 F20101220_AACRPC li_y_Page_048thm.jpg
0c9ebee1f9539d6103a88a57b4c6cad0
cfff3a3683ef0ab533f1ea85922a8588c33a1100
F20101220_AACQMA li_y_Page_059.tif
17a9c47b1dd1f00305457d07ab8f1f8c
d1aa60d174ceffd7b1025b19ffc06adae7e5675f
29963 F20101220_AACQLL li_y_Page_012.pro
01a457150735dc6ebcbd3ad242fc51a2
2cf7c54a11ad471e011b3b1d1238fa7a90f5e49a
23462 F20101220_AACRON li_y_Page_063.QC.jpg
34938de13d678058eaeeb24defa6eddd
71f993595f2304bf011303fc43538eda72dda9b8
1212 F20101220_AACQKX li_y_Page_078.txt
0a64f6459587de8150a2ecb41b64e11e
8e43a454bfda13fa04da46c08bf89086150775d6
67758 F20101220_AACRNZ li_y_Page_145.pro
97279dc1c46b81065ae3c3be88e39838
7e5cba1c77ef3ca092617815602e21f3e0adf3b2
37261 F20101220_AACQLM li_y_Page_071.pro
32df5d13899dcb455fc1776c69304b6f
f95934c0dd5cf609c5d5d0e7c46904db88c7a7f2
86986 F20101220_AACROO li_y_Page_007.pro
5b4388018c3667ece59168d732c8a1ae
acdd66b9e7a107819e86a081db5750a138677fc4
55495 F20101220_AACQKY li_y_Page_042.pro
e356e849c338e439acce7bc471153de9
22df96c16b14eaa2ae6739c59043edad70e6753d
99249 F20101220_AACRPD li_y_Page_033.jpg
76401401556f7d4c6c16dc57176ba446
e13fa9d866ba721827bbd7f950fed9861743648f
F20101220_AACQMB li_y_Page_127.tif
4e3c5478ee77bee6853226719f75d222
0b875356b97a204277524d8b6b7bb945e37e29e5
154743 F20101220_AACQLN li_y_Page_112.jp2
88311e037864fef2d2803187360498d6
bd2b7b1e3f36dcb400888090e05cc523a2eeebf5
44192 F20101220_AACROP li_y_Page_040.pro
e617f6826860748d45f7770fce275165
d4829399cfacf4f17be2b85942c51a0d9df1961a
28282 F20101220_AACQKZ li_y_Page_092.QC.jpg
5303db4c0fa11db9d18cd90c612a991c
49e60399ce025f8dc321e2e79e2edf977353f132
953301 F20101220_AACRPE li_y_Page_092.jp2
24086f9f42f659a748c5ba425546126b
54ed477b8b55b7652cf59870bf1db49b8015f836
35273 F20101220_AACQMC li_y_Page_141.QC.jpg
44719a3ceddf7834e253fbec69d5a2fb
276b61679a0041bdb06dfc8ec0cb722dad7b36ce
604006 F20101220_AACQLO li_y_Page_132.jp2
817f4221cb3241ca4763b6a7b9428059
e3259eec1b8b2ed490eb893821b503f1b51071d9
F20101220_AACROQ li_y_Page_125.tif
9e0bf9afb1a26c879a5b554863601c3f
7eb5f844cbdf9246dd8f380ee4bee39079c86645
9241 F20101220_AACRPF li_y_Page_147thm.jpg
6451e8747f01acfc32398e3393e191e2
c16f694ec9818c85c3af1209358a6b34d92eab1d
1385 F20101220_AACQMD li_y_Page_136.txt
05e08e43d1733ad143b8b3476a7a8cf0
503e6c4cbf35616aa1a0fc8c5de18f97d46fc644
1980 F20101220_AACQLP li_y_Page_118.txt
cff25cd9da8946b9bcc3482e65503daa
2e104efaea1a8293ee91d84b041365a67314d190
F20101220_AACROR li_y_Page_003.tif
36ff3bc036a7290042aca994e9d421e0
91e8ed074bf7fe4274425826467cda994e838115
17707 F20101220_AACRPG li_y_Page_041.QC.jpg
2bec33189443908ea3c49e91c2e56af7
960ec91c38f334eaf123c5dbdb603b0266f24345
19387 F20101220_AACQME li_y_Page_120.pro
6ff51ea08dda76a155ff5f68c2ebec84
67492521f548b4a1e7d2f4761c47a2e2dbccaf62
1113 F20101220_AACQLQ li_y_Page_125.txt
4755533c6442dc2fb47c73530de8be2f
af807893013cb9a02dfed1b2a89ae4d447401d60
87134 F20101220_AACROS li_y_Page_060.jpg
585e43fd1491f0312514b3d913dff835
eda109e8a0f9923e99eec2e07b28ada7c6ea8043
738 F20101220_AACRPH li_y_Page_128.txt
cb9bf93e99ea3cc7b2da69291d1765f8
67085fd67adc9e07634735ec995b04da603f3558
1751 F20101220_AACQMF li_y_Page_095.txt
2cb21e5bf58923df7bc3109ccc0c39d6
c55aa2309d437be639a1aa91f3df0a30f14dcabc
6495 F20101220_AACQLR li_y_Page_025thm.jpg
06b4c5f946fce9efb76ad2487cb7e0c4
9d50a1ec2270e91cb146dbca45e14a0a3c53fb1e
F20101220_AACROT li_y_Page_077.tif
ff885406d583872f2c57b5e91918c40c
18239f8c468ad074c84a23a15671bb49afb8a588
70291 F20101220_AACRPI li_y_Page_151.pro
46ed9245c6c39231deba037e47b18bf2
479333f580b02401f14b4f376efea87c2a4103dd
63035 F20101220_AACQMG li_y_Page_130.jpg
b1f7a80a237b49bdd1641212d5f71cc4
6c862109bb171235304441157f358ed63eeadd6e
18564 F20101220_AACQLS li_y_Page_044.QC.jpg
bbff7325456f6d8d6971fbcad0c47429
38d7e58797abf5e300b34b5d5d2c2c6d56a17660
F20101220_AACROU li_y_Page_006.tif
bc8b764781fb2bc377fd282a2e118827
b79067514d65d9827354a3352683f1fba089115b
105343 F20101220_AACRPJ li_y_Page_055.jpg
fca83179fde491c9f5e9736c48a0cf89
aab0403f7dc17562e3b77bb4d99e7c42104c7842
107150 F20101220_AACQMH li_y_Page_139.jpg
c4b5ee513a6e97d36874b67c4a751924
de5d9b2408185267b7f0096a5b423bd08000c757
F20101220_AACQLT li_y_Page_065.tif
0364c5e698cd87f25292bfcb2fd95d8f
7c7a033e6ed671b52c6143666b52e6573bd5d5b0
1768 F20101220_AACROV li_y_Page_106.txt
c7f8fefed3a7858d849c4e36964cb3ad
6c71a48b82150f230fe1016f4d8ff531d4a0ff3e
75729 F20101220_AACRPK li_y_Page_005.pro
cc377f64cd7d51fa6fa14bf518693169
ccde86c7801646d393214679fc64955f1ecb8940
5896 F20101220_AACQMI li_y_Page_045thm.jpg
36de8bb775239b72d51e067d743104e8
9745ab0a7c696e6f09ac95cd877645bdf28f84b1
1911 F20101220_AACQLU li_y_Page_084.txt
fe8f50333b62ff34d32133325ad5672f
83d7a0150664b99e77a3660c3ffe07f08daa3f65
F20101220_AACROW li_y_Page_047.tif
317a2ce07757146fd5c316dd27f7d66b
0f00d712972c6b91891968cb5d9447634ca2f251
60345 F20101220_AACRQA li_y_Page_008.pro
e66c0a9da6ddd35b831be6cab067d1e4
a172e58b8b894f4ebf1baa35054d025c9fd3fe07
F20101220_AACRPL li_y_Page_057.tif
78d4b1db63c89544bc0f324b063470c5
ec869a41bc65c7aa7ab109d2543fd0745dc00559
9253 F20101220_AACQLV li_y_Page_140.QC.jpg
b0303c097ec8e07bc8c7f54d8b36afff
9642c971c04c76a84d052272895480f1f70eaf01
F20101220_AACROX li_y_Page_005.tif
f72dc2ff4846411048c190f506f62ed2
4898c524ac8d3cb9f48d3bc610d55e715cbb15bd
31135 F20101220_AACQMJ li_y_Page_110.QC.jpg
d5c139690138804d2a4933dcb6e0bfa4
5473bd5030e7d3fec38b00ba20a120a28113362c
4518 F20101220_AACRQB li_y_Page_012thm.jpg
b4572f226c4316125dd248545467e8f2
9665cf1763788673ba9b21b66625c8f5686b2b04
29080 F20101220_AACRPM li_y_Page_026.pro
9fb80c1d3f267b09f4eeb3fdcfc175ea
9947727a49e5b3431abf63fe0912db5821c0e28e
12978 F20101220_AACQLW li_y_Page_124.pro
613f5a7a3c0e4f44d3844746aca937cb
8b650d5f935dcca8e6988db902681acb3a4a76d1
41040 F20101220_AACROY li_y_Page_052.pro
482655ada4b417478ef43a14073c9682
c00b92cc52ff57c2b30822c625563989bc6c0752
104530 F20101220_AACQMK li_y_Page_108.jpg
d5e712ed23fb75a68f3b6111c6ee3253
8a7e452d9203fbc9f7f026ccba84a575adb08575
437116 F20101220_AACRQC li_y_Page_058.jp2
fee6a088d009039fcb62ca4de4d4a200
a0eaaef8b92985f837612ca35e04b37042308c48
41201 F20101220_AACRPN li_y_Page_095.pro
a1c9a3f4cc4a6262f9bd11e663c2d5da
6b3df0222f9edb45967e728a7fb24a3d64d81777
F20101220_AACQLX li_y_Page_139.jp2
1b41398d06f60d473559f93bf22bbc2e
cc9947f76d7829a93968c3588392030d63df3d6b
7223 F20101220_AACROZ li_y_Page_100thm.jpg
d66dbc8030629f55446309ccf9cd2947
f35cd9b463ff59eb63b4ffc5a27f7ab52c0acf09
96371 F20101220_AACQNA li_y_Page_034.jpg
a579add7d29ac291541e82733c97b676
dfd474d5cba11dc4f11e03ed31e2e1ddd130c762
1012869 F20101220_AACQML li_y_Page_013.jp2
e9ea9e6a718deadf16410656d9749509
f747258bcff5b0ae15fbf9826e90bfbc143b7614
5344 F20101220_AACRQD li_y_Page_129thm.jpg
489f5bef9369e83c741a199e8b9e3bc7
18df496621cc25833f5644821d627bfb4a1b00cf
6031 F20101220_AACRPO li_y_Page_135thm.jpg
76e893a3a64d0ad048ebe5d8daabff99
929f777fbee45f6602cf8ce634fa247ea261dd18
69428 F20101220_AACQLY li_y_Page_144.pro
5dc0614e8b73915fd0859f49b61e6e16
b75c072e130d41aec11553377416fe554da8c4ad
F20101220_AACQNB li_y_Page_033.tif
a3dfe4bd2e5dafe8134de2ba4988537f
4e3aadaf9be08d816b7427c2c4c2e34561e68331
5658 F20101220_AACQMM li_y_Page_127thm.jpg
1895961ffb465c0d6609305eb41f18ea
086816288f20fe8dc07542aecb088571d23d97af
1605 F20101220_AACRPP li_y_Page_004.txt
ce23cbb821007a8a0fddbe9e3d2f11e6
3e37aa709c2915e178bb03b4c4f65dbd5efc5c00
2776 F20101220_AACQLZ li_y_Page_027thm.jpg
ef25c2d7811b99f75b077779d6547da8
2d22d974120c7ab14900e50a4c6df5ecdd485ad9
7869 F20101220_AACQMN li_y_Page_110thm.jpg
97561a44a360635939d6f76dd63933d2
b972b46bb1c7944b892605ac6d8606bc28ce4230
38192 F20101220_AACRQE li_y_Page_144.QC.jpg
94e5920db3058e8ac738ccc0d5802850
6cb02f5607a6d47c7e85fe122afcc5f78816b210
101718 F20101220_AACRPQ li_y_Page_050.jpg
cfa8524b3aefd086e3a9c4fb9c998181
3a7218e26087405350defd8b80c95bf7b748a3ff
6533 F20101220_AACQNC li_y_Page_037thm.jpg
7943a0adb6847a248430f3a60beec251
b859407e0e73c87a056cf819b4c1810d788b036e
28813 F20101220_AACQMO li_y_Page_082.QC.jpg
596fa5163eb4a0fe19ce75d1c4c168bb
29714c70aea896022f0e6f34c1b9fe1503e04768
227335 F20101220_AACRQF UFE0022512_00001.xml FULL
966ad616f5181477efd183e7e1712c68
8f88a3c7acc19402ba1e1b0630928a57ef58d000
59013 F20101220_AACRPR li_y_Page_124.jpg
ff33360d42636d1dd78e1eb3ca81e1ea
2a0f34cbcc3f0e78162e2d2ea289b3e777130a9e
1051938 F20101220_AACQND li_y_Page_084.jp2
c2c5c077cceef0ae08aee0e513c3dac5
299aab3cf2cfb5701ae816317be326a20eceafba
47461 F20101220_AACQMP li_y_Page_093.pro
ffbfe393cccb18b01e2a2b0a131837b2
dfb32c4a9e6b0b9da8471cd766356a5a8c64752c
8080 F20101220_AACRPS li_y_Page_016thm.jpg
72f7b7b2660157e1650ffc0ea0820ec9
3997cdd93ca012787fc1dea38c62c5244510ba1e
1226 F20101220_AACQNE li_y_Page_129.txt
4f2bcfac65bc2b55520dbd46d3b4d4e0
58293ac46172c83fcab5ff841e10eb13255420fe
1051924 F20101220_AACQMQ li_y_Page_017.jp2
7a039e167d3329d6d22dda24092b5411
33665c064516215645e5ba5be75c467962d0695b
8551 F20101220_AACRPT li_y_Page_067thm.jpg
0aca50b9ba81febbeee8187ecb3210a1
d20da577f98f2f6c3c2be0d1a076298085195d31
9298 F20101220_AACQNF li_y_Page_148thm.jpg
e5c014163f126fa18bdb230816be046e
85c7e4cf08a0cea4fbf86958e3c5458e3e3b4d43
99263 F20101220_AACQMR li_y_Page_061.jpg
7ad5e5b3fb11c49588d1c28595c85edb
cb8b676d5f43e79dce2944494d76c381430116b1
91701 F20101220_AACRQI li_y_Page_013.jpg
91e18c0906f7017f3f45882951568920
95de5f096e4ade73a933fcb8f46ddc70622308f5
140536 F20101220_AACRPU li_y_Page_152.jpg
7bdd66d9cce7a5d19139055084a74c7d
57420c7cd38a987c55d29da431eb2b2901d3c88b
402 F20101220_AACQNG li_y_Page_045.txt
f4c92c2c2648f9682039b20b37da8971
43eaabad7ecbfcd1728dbe18e412fc7dfba327fc
1051957 F20101220_AACQMS li_y_Page_108.jp2
4e59804d295bd154f096d700244604b7
9bd90cb399f3d5b4ca97278f6b2e1c7a2e676bff
108677 F20101220_AACRQJ li_y_Page_022.jpg
6bbfa22bd48f0bdca012cfbe2b4cc955
2277c92c3c71f4997ae12ff2296a00cae672bbdf
127970 F20101220_AACRPV li_y_Page_005.jpg
7ce5f576ecc008603b73c07d7f3218fc
5c03b301fefe3b8cb0831f50454081adc827e42a
F20101220_AACQNH li_y_Page_096.tif
4bd9b2fa409165a50e53d35458d08e88
8117c4fa4ec968b7f74dae9295c3cc4022f6e0ff
540242 F20101220_AACQMT li_y_Page_038.jp2
72b7653d02fd52c36c4f5c06fa0980f7
fa2de81b03ba390cd4254fa59727030a7b7d5bb0
117526 F20101220_AACRQK li_y_Page_023.jpg
975f83c87c291f4ea3f5bb2bed2fe5a2
2efea29d576cc801895cec92d4350f96655c6b01
7377 F20101220_AACRPW li_y_Page_001.QC.jpg
3c5fa63cce84ba635bae3e8b3c20b0e7
f4874c21ea3d7acc412da99666d9b589d49708e0
52517 F20101220_AACQNI li_y_Page_114.jpg
953afdfb3ba21305db39f983c509df15
d4253984b530f604e416f7687c7810168e0e0bd4
56932 F20101220_AACQMU li_y_Page_020.pro
56fb9f12057a5101fc6e80fa62c445f9
d81aba6ff89121fb31d638b0099f7765d063b235
63217 F20101220_AACRRA li_y_Page_087.jpg
d12b68b2b983444e9357b99d615e259b
4a5929f7aae049586a74eec43ba59c88a46ac518
77094 F20101220_AACRQL li_y_Page_025.jpg
e509274d6f2e7f644c9d12366de3902a
b401c5c0b052933364757f1bf43231541b3b774f
25939 F20101220_AACRPX li_y_Page_025.QC.jpg
d196dd08c9a07c6d807489d2c920052b
0a38b794ec4f3e4c7c092f872f338797f1f1addd
71506 F20101220_AACQNJ li_y_Page_046.jpg
b0898ef199e8ac051a4a32000723d027
1bab28d52f847f75d90a5e6f01f0ce31aba81ab5
18265 F20101220_AACQMV li_y_Page_039.QC.jpg
0c06fec5f09f36a565e24f0cb2552c28
d171cfe04f4249f530badc7c6e4b42a4e294e427
47995 F20101220_AACRRB li_y_Page_090.jpg
ee71e9c9fcfb92da383a88de6f3586f6
b2d63800611a7e2be3ba4a28408f7ac98e6019e7
61779 F20101220_AACRQM li_y_Page_026.jpg
f39359a346563557db0f396ff21369f4
ad7872242d4f3903e751c5e7fa1ba13b146b7f4d
11307 F20101220_AACRPY li_y_Page_088.QC.jpg
73d1f1b325a0bb47ed00c928ca6eb160
b1487ad165cf8b7623ff386a537b50fb9016b92e
8249 F20101220_AACQNK li_y_Page_121.pro
60172797b93ad7dd36aa3a3d753489cd
919d476b9f2b597e62aa0ca060e7466f8ddfa589
28030 F20101220_AACQMW li_y_Page_155.jpg
6b7476c19528db478fd0853d66b9391f
2fa257370008c41e94e075ec62bcfb02cbff423a
88395 F20101220_AACRRC li_y_Page_091.jpg
5b115840e27f6e4ffe4d62e9078f54fd
b4c32446a167961a3b97242f72d2a665e4ecc668
32254 F20101220_AACRQN li_y_Page_027.jpg
f06fcb54c0a63a23b3ba2e7a48b8e70d
0d746db71acd17a3ab6a317e6446d8cbcf3a4ded
2837 F20101220_AACRPZ li_y_Page_153.txt
592b5948a015a567268eca91da250cc4
e34fd115c5e77e01a27b1e14d3c787768c9e769f
105368 F20101220_AACQOA li_y_Page_070.jpg
9b1f65c2cd30521eeec1435315860009
328ee2344005ab9bda9e64f2a6e59b3694ae3d44
138326 F20101220_AACQNL li_y_Page_153.jpg
af2a3e15dd2c1c82a0ed13502c0dfbb8
ce76af0324ce1bae1a1e4e18ec0895e17d729a0e
7515 F20101220_AACQMX li_y_Page_082thm.jpg
eaef65d23249610ec7a780d7fd334a78
f3eb15bb5ee553bc1ee4d751566ed54351cf0456
93225 F20101220_AACRRD li_y_Page_094.jpg
587cde8e7ff99858bf06ea744e40e51b
9fc0e1547af95bf9bd5647fe0549a8e370d6eef6
102873 F20101220_AACRQO li_y_Page_035.jpg
ffaadd5c10aeb66fc1a1681c2626bfdb
f21876d145c3fbda0f218a74a211bbf6f1d091e1
36483 F20101220_AACQOB li_y_Page_025.pro
e72b187b309c606afa55f1c8afb199f0
6212a0f939e48944fb8c021d56c09db7dc5473d8
27686 F20101220_AACQNM li_y_Page_056.pro
f4c583f8806e960dce818b2724167f83
fbcec0fb45d9459f97afb3ffa4682e31de7248ee
66199 F20101220_AACQMY li_y_Page_143.pro
39a3399ca101a1506ac44505a676366b
7ba432e05be63bc2e6b27c9cc80a4c82918985b1
84425 F20101220_AACRRE li_y_Page_095.jpg
a2ac945095792a2776f558058239835c
9f7e739c90e415c68141b7c5acb88503ab573e2a
83839 F20101220_AACRQP li_y_Page_052.jpg
97be6944cfb85c8f0d0969e1ee295238
5f74013ab379eb8b82502a9059292217b65c3c5a
3128 F20101220_AACQOC li_y_Page_029thm.jpg
affcc25a833d9b4356e012c7e016ab1a
bdb72287df9322f2823b29b177426e136779e095
1565 F20101220_AACQNN li_y_Page_071.txt
2eb2b707bdc83a9b359a2caebbd992b2
fdb5907ac0630695a3b6fa798241e6d62660a9b4
8322 F20101220_AACQMZ li_y_Page_031thm.jpg
b44bcefbfc01be117ffdda234a59bf09
e7b1f5198bf04078ebc6685494c24d18c30e8ffd
57675 F20101220_AACRQQ li_y_Page_059.jpg
c412cb937451f51f789902219ddba296
a1b6cd7dbe4593e3a88bc0b2c3cab1f5b26d5b17
49755 F20101220_AACQNO li_y_Page_016.pro
8434ad7d9a11ffdbb937688cc3944ca0
6b767e63401633317f34564be5e76adb4da9266a
103728 F20101220_AACRRF li_y_Page_096.jpg
f810e8521e965cba32520cf4a9ecb82f
d3e804f65dfd72ef75c2c6133d86fe32e1e2398e
109654 F20101220_AACRQR li_y_Page_066.jpg
52b74d543540850818da78a1ba610e9e
c2d076b9dc7afb01b3c61dff087a3d4bc61b0c99
8545 F20101220_AACQOD li_y_Page_062thm.jpg
747f43c76d138853545952568a58a564
444e7c9f0a6ce1e5b7871229a66d4e3c1199502d
F20101220_AACQNP li_y_Page_078.tif
1b02b3dd01a612fd3d7269fefa179116
47434d56dc65605937963680a9aadb7c60f07f5d
61995 F20101220_AACRRG li_y_Page_099.jpg
ba4128e0ca8cd6dc2297cbe0e1c52f2c
d67fa286be4f1da5f75b81ae602ce3615adba96c
108555 F20101220_AACRQS li_y_Page_068.jpg
6b679810f956de9d4e6baf2ec0e6da45
026edcceca292a3146b116e83426fa825851affe
35100 F20101220_AACQOE li_y_Page_008.QC.jpg
32ed81e768af9b755329978a3d45c830
99e3f65d886fa77fe358732a6b71fc2519d60cc7
57143 F20101220_AACQNQ li_y_Page_073.jpg
b7b4777a97b16f72f158df5400053091
fda5f1bd43c89deaf1304dcd7892926cd60e4647
95894 F20101220_AACRRH li_y_Page_101.jpg
b19e2428857f80b9b71a66c9a462097b
ac973520a927f6d9259cec33bff0ae97e4991cc1
58792 F20101220_AACRQT li_y_Page_071.jpg
f5390ca3931b7be5ca70ca381be63e8a
8b1f60d3fe3facd1a2b32e9ddbf4bc424121b4b9
F20101220_AACQOF li_y_Page_120.tif
48d3a2e0c59519f79e2e3016af130a25
656767f99a5487c81acc61c819b1c7573d9f6d59
F20101220_AACQNR li_y_Page_069.tif
8de9779907dd8bc237d17e634d3872f6
b03f33e13b1b3263a114c88fe422e6222c23d4c0
97450 F20101220_AACRRI li_y_Page_103.jpg
4013a19b919cc7496bf3da39cf69f29d
30f089d18b1d95deb0df0388efe78687999f74d7
40012 F20101220_AACRQU li_y_Page_076.jpg
ada3c1e39a6404fcc76070c46fc0d3b2
b0942059b5b2e5d87f5a06c21b91def487c9be18
610130 F20101220_AACQOG li_y_Page_123.jp2
4ea0efa42cb434ebab23dc126a4a9820
fe019d3085789b84c4c3f015d3f8fd376e9ed8b9
6257 F20101220_AACQNS li_y_Page_080thm.jpg
042bf8b8865a6cd54e9eb6873bcddfd5
e8889c7cbe247f1657f331fbd99da7b94e6b578d
102485 F20101220_AACRRJ li_y_Page_107.jpg
043b759ce3fda6c9c9ad9d50f16a97af
003db0732eb206f366045949e057f3b95a065446
69973 F20101220_AACRQV li_y_Page_079.jpg
2ed3f06720e575fdd9a4ffa168505934
52e01dd27c6b96408e0559770aa9428d752e1c9a
1850 F20101220_AACQOH li_y_Page_081.txt
999efcf4bbbed4416a6ca736b395a8dc
9e38f82b01b8bc0df36ffba0862ecb5c524ca44e
2843 F20101220_AACQNT li_y_Page_142.txt
63fe66483b214251ce7c2022e39e6f7f
3c41a186ff038fbd90995d2f6292c2608d2980be
17615 F20101220_AACRRK li_y_Page_112.jpg
53be7e78a3561076dc7775837da6d4c4
438c09e5f683a9273237817e16b05b40dca25861
84488 F20101220_AACRQW li_y_Page_080.jpg
d03262cd05c0a52ec22bf475c0aa5aab
5ded3e46f765fda9d1286c5afdef5a840c67037e
F20101220_AACQOI li_y_Page_155.tif
2182327afc5c7e83546fbfcb0a7a66ca
cf37f4ba57b82eea4dc1d698e2ed53d3c69e0b0e
61671 F20101220_AACQNU li_y_Page_131.jpg
9402d7f00ba346d330876e978041c368
3e8d91614c5a7bd45f7cbcfa7fcd8024ac93c348
732333 F20101220_AACRSA li_y_Page_014.jp2
af897df240918fa7b0b192f65a147c6e
11a751d52f8716822a7a88e43aa5d0a075815540
52371 F20101220_AACRRL li_y_Page_113.jpg
8f9f6fc7303598b6cc8530ed5fb5630e
2827f601153991a2ab4d8c83c7417b18217d30d9
84494 F20101220_AACRQX li_y_Page_082.jpg
242a88234111498a698004ea1c96d7e2
a8577e27f52eed81669ac6619d993fe0af819fb4
8796 F20101220_AACQOJ li_y_Page_049.QC.jpg
c680957815a3249fafe156cd6864352a
92aa372d311e7c5b65029eeb84a08458fdc0220a
135394 F20101220_AACQNV li_y_Page_145.jpg
5a08fca6cb542f7556c87588f69b5dd3
61b6c82304c2a8f494ab39ca054d990e1cc24ba5
F20101220_AACRSB li_y_Page_018.jp2
e97977342a11a4db31cbf893204fe219
62b4cb9221e26cfc84cac14e7f4617dca0972787
55678 F20101220_AACRRM li_y_Page_116.jpg
6bdfb5230841b9d4c1a34a8fa33dc847
3132f8ccda46c165a9943e3c9008ce84aaf81e85
90396 F20101220_AACRQY li_y_Page_085.jpg
6a1256dd95075851ceab53919a479334
9555dc438b4d9787f73e135afd4481f1cc442bae
699 F20101220_AACQOK li_y_Page_048.txt
2c3aef064caae868d293fd972a530a84
deabeef818209c7fabc0e114034e2db0b8becb69
2804 F20101220_AACQNW li_y_Page_152.txt
b41156b18f3555bc0ce1ed2f7d4fe23e
fea1ff3685275a6df90b21a19bdef7644eb8b8fd
F20101220_AACRSC li_y_Page_020.jp2
14207c72206f67eddd5fffcf197ae226
4224a35fd93aa768584cf4ae99a5020343ec89e0
86093 F20101220_AACRRN li_y_Page_118.jpg
cc5048ad8ce73acce629177a08187a06
c5f83b956d6d034585442888e86f092e970b89a3
92072 F20101220_AACRQZ li_y_Page_086.jpg
c0ca34053043169212b6d00f8b42fe15
fe7e403005f55cf9ad8f7055d8e306f328989155
33591 F20101220_AACQOL li_y_Page_017.QC.jpg
dda64e16f2e0d3fdee51fde3c263d00b
d5fcdde502832e48856310a01fd841f192a754c6
128372 F20101220_AACQNX li_y_Page_146.jpg
ac1eb1cc78e59765353ab4592eb803f4
345b22fe3d57d20859bf9e3db4e116eedf96d115
54031 F20101220_AACQPA li_y_Page_069.pro
d772b45489560ee61a589684fdc117a0
d79006b42513a93a0e4c93fa44fb0fa6e303d378
1051964 F20101220_AACRSD li_y_Page_032.jp2
7e8570c4723840e33910dbe16d613dbb
2eca78402aed780f62bfa229a5777545437bd674
25423 F20101220_AACRRO li_y_Page_121.jpg
0aaf17b90837efa9bc43f4cb4f05f897
a52ece0d779de2a0c2daef7c2afbebfadc302b98
63279 F20101220_AACQOM li_y_Page_012.jpg
c2c76142ee1b131219af94d723d6855d
dd0c762810a9edf2277367ae60ff656481e751cf
17457 F20101220_AACQNY li_y_Page_089.pro
94bc01f33e0d30677f44423d8bc17b75
2ff98ddd31abfbd7f134bd768043d339f7a13dfc
F20101220_AACQPB li_y_Page_139.tif
3a0a562184484501f0f71faf26e4c176
2a11112856cd0e79889f27500a3252de0799f490
1051959 F20101220_AACRSE li_y_Page_033.jp2
3686dedfbc648f90f42627454085ca7d
c93bd1a2a4d21e74201897c88031a409babe74af
60599 F20101220_AACRRP li_y_Page_125.jpg
228473d700cc32177d50b03bd007890d
8a81a96b7f277fb832063a8990cc77f3cdda9ab5
1051976 F20101220_AACQON li_y_Page_093.jp2
c065502e3fde2351acf134515d78475e
07a735f14879f29e8c27fa8aa925e85ad94e7d0e
1051947 F20101220_AACQNZ li_y_Page_055.jp2
5c1883df4f770d9306110c521b047b08
ba29148858d3955286c51038931683e0f4f78706
643202 F20101220_AACQPC li_y_Page_076.jp2
7d96a1962239dc13ba7997833303415f
0a8e2f2718c8571f7bf14fac69ac66ae3788e205
911192 F20101220_AACRSF li_y_Page_052.jp2
3b8ee460db3127c162a7253521eab8a1
888d03be71591d4f0999c76d7e83f301380db876
56248 F20101220_AACRRQ li_y_Page_126.jpg
1f63875a98b0f8e76867846e8913dd30
a068878deceb18d931c56f2aac66527a44a81c82
F20101220_AACQOO li_y_Page_104.tif
96aebd9da43448101f7fa9df36b5b2b7
d40d13180751a695dfc91d99213c25d63fea38d0
4138 F20101220_AACQPD li_y_Page_038thm.jpg
0709054e9ea16e100fdc5ead05c8bb4b
91ea7fbdebca3dd3c64af4ff38882feb1256a488
59729 F20101220_AACRRR li_y_Page_128.jpg
0e067aef34ae6008533eab7c7b7420d5
66a5c85039ce2693f5ea0ee5740d7ccd5960b547
29201 F20101220_AACQOP li_y_Page_085.QC.jpg
9c880fc16f8dad9a3eb195b316dcee53
ad113fb741be2b109e2cf7dff501135827927f24
1051984 F20101220_AACRSG li_y_Page_054.jp2
4b2f2f8696bc7b34b3b94b5ebb828520
c54ba6ef47a97f7df0d92d76507cb30301bd5596
43546 F20101220_AACRRS li_y_Page_133.jpg
67ed426d18858579f1c345716ef8f0da
fb86a67394ebf1dd1433ac0ad8e0565a16094031
100090 F20101220_AACQOQ li_y_Page_109.jpg
ec11bc6bb35d32f54cacc1119327f892
0c39194be26bef33d983b38f4f34cfbe8671c348
1051979 F20101220_AACQPE li_y_Page_150.jp2
f376f9ab3ed68eefd08e877c74e4d99c
d144c6102a2625b23258c22f0f4ad08778c73e04
541766 F20101220_AACRSH li_y_Page_059.jp2
f0db5693e2622deb67fd857764382a35
162eb76153749d934648789c867d6d5bdeaff598
45147 F20101220_AACRRT li_y_Page_136.jpg
045d69ae301a2a2fc1b62a63ede1dccf
d6e59f4ea0fcd91e10896cca79b2391a372b267b
8487 F20101220_AACQOR li_y_Page_105thm.jpg
f837a40dc266d9e168708f3c7b07bc43
11f58c514f3973c74883ee0d6276327f57a6c3ed
74783 F20101220_AACQPF li_y_Page_134.jpg
1ad20e77454801050372ddc81dc2b568
58ea1d254bff00f96108b143447f406a50988c50
1051986 F20101220_AACRSI li_y_Page_066.jp2
ff7b1145dd194442d83a45606666f83a
e8a24e2c143f33d011c0a0528a000577dc237d21
132115 F20101220_AACRRU li_y_Page_143.jpg
0f31440faad55774945871e19d1b37ab
01d2df11040c6b06e9141dbea2decd7c1d37d817
42307 F20101220_AACQOS li_y_Page_132.jpg
ad639127c8f4143b79b6f47901ee2ab7
43b5a1637a06e6754449e854324f8fe80cdad437
27557 F20101220_AACQPG li_y_Page_005.QC.jpg
45b69c89928784df276b6eeb183b080a
97b08a68ee30a6015a4a7297edc35731268ee178
1051908 F20101220_AACRSJ li_y_Page_067.jp2
709c798378e082c90e6ed77d85c070b7
39c465b9ddb4dd7c13617e96c1fb55af57449663
129044 F20101220_AACRRV li_y_Page_149.jpg
4940db8e8991c9f8a78615f0e0c04a2b
4c2fd47c13d2923dd0532f98bd9a5af259f777fa
39614 F20101220_AACQOT li_y_Page_037.pro
cabed7b04aeb977de830903256a81b99
b94ef953ba5aece6ac8eb1c4a1f41cd06486a194
32784 F20101220_AACQPH li_y_Page_006.QC.jpg
bb041a7fb0269efaecbd54e35607d801
75a045ccd08691c362a3e8c1cacf700b281bf5e1
1051982 F20101220_AACRSK li_y_Page_070.jp2
45ab15ccd63f6dfd2f64d61270eb0538
ad3db5c608f7cd58c9cfc3b835908720a9168e1e
48090 F20101220_AACRRW li_y_Page_003.jp2
347ffc02e6317f0608e1db380fbab0bf
acbc8f4298c8813bf3202b2189dd1b32fe8b52b8
49186 F20101220_AACQOU li_y_Page_031.pro
61e00597aa28daf8bfed4b31e8cc851f
abe05df6185ed6950bb2d916343c71f3a78e0066
61039 F20101220_AACQPI li_y_Page_141.pro
0fb6c9c1cf730f0d6146c6ab44623bfc
2546d6adbd389d3169de4c7ea1664b53ed64b8f2
F20101220_AACRTA li_y_Page_141.jp2
f1af4d74314211b207126931686636fb
18b3d450868bc684fe8a00ad00b6cc3e86e9976e
903776 F20101220_AACRSL li_y_Page_071.jp2
a5dad28859c6447a50375bc5a51f44cc
e3626951b83c5d193a4c44085960289e871cc786
F20101220_AACRRX li_y_Page_009.jp2
b325dd9d66159c89f9682826f045a9d9
12a2fa3f2d8fae6f161d1dff4901b40f0206471f
34659 F20101220_AACQOV li_y_Page_070.QC.jpg
282fbaa2011393aef094d00884a04f1d
38baa8d0be80caf3341df227060e227b416231ed
73039 F20101220_AACQPJ li_y_Page_042.jpg
ae4b2175258dbbffdd548913d58c5b2a
a9796584d1d6ef2c0cddaf6580842d8fdc58bd3e
1051985 F20101220_AACRTB li_y_Page_147.jp2
b55ae731e1709fe0d4ee187870fa5a08
03d9c3f9f8c57d84e85f19106d57fd4ee1a2f240
900499 F20101220_AACRSM li_y_Page_072.jp2
7460cddf408c70c13f5a1a1348da425c
7ad651394e60d3d1bd177b0a2cc94464b01ee9ba
F20101220_AACRRY li_y_Page_011.jp2
e442406f89e54574ed70daf702f30fe1
09e641c4e0fe2591186a75c91d9c309250d24712
F20101220_AACQOW li_y_Page_039.jp2
32214f9c11779dcac784a11b77b39054
5da3a23d76e839dc13364bc30284f97e46e23f60
1924 F20101220_AACQPK li_y_Page_097.txt
9e01233ca4340adcc588e2805b845b7c
0fed9e56ca3622a0618aed1f901a0d48f1ceeae6
1051966 F20101220_AACRTC li_y_Page_148.jp2
9362c86bedb6ca09fad29924fe635940
f6be6dd7510561c2bfaabd7b05938d3f61334160
892560 F20101220_AACRSN li_y_Page_079.jp2
5270e70e2bfc0cb0f14b2f0b76844a24
04ab269387d2a70dc50d2b5021a9d86ebf007396
F20101220_AACRRZ li_y_Page_012.jp2
7ab3b5ecac1e7b59997fc7f332b50207
bd9aa42cb0e1da5bce367c14d8689cdacf0867c8
30829 F20101220_AACQOX li_y_Page_101.QC.jpg
5a4b2d8b62fa413d3c8342bb17601223
712a9c290ea3089fbdcc66108e7edfb1cc549bc8
26154 F20101220_AACQQA li_y_Page_037.QC.jpg
d5c52fd400798047c7bdeba3f44c91fd
aee3af2fb22f140b7e17f3eb19ca972cbad41f16
2949 F20101220_AACQPL li_y_Page_075thm.jpg
ca429f711f20f346cd302e4f381083e2
9eed1dedb054617d7cbfb19c5f2a8a348aa4c1ba
1051962 F20101220_AACRTD li_y_Page_151.jp2
fbbc5d11fed1df956697070dc6186e69
cc1fe515b61a01c52241ec51b2005d5b30d40220
983427 F20101220_AACRSO li_y_Page_081.jp2
5939f77874e55d98255e0a7721f723ac
24feb4942ec3dbec276e0733ce9731b561eb975f
430983 F20101220_AACQOY li_y_Page_089.jp2
b52286dfc446149b62558463a24d48f4
28e507b522cf301ffb6f7e57bea9bda7c4073792
35629 F20101220_AACQQB li_y_Page_043.pro
123020a642939976882ea63915d798b9
123b3da2f4157e3018788fd96a5b6064f21348b5
2144 F20101220_AACQPM li_y_Page_024.txt
238d636f7a78025f1fefeded6a69094a
d95dd109e752adb035e7b59385db037922a4e756
198316 F20101220_AACRTE li_y_Page_154.jp2
2d884bf87829591ac76a36155799cb05
19cc75e952ea894332a75632482c379be2465776
459941 F20101220_AACRSP li_y_Page_090.jp2
be92c54e1e8c98727a86d6ec11547e51
c173a74701d7ca2cfa3e0c70a731cda2bf3ac497
F20101220_AACQOZ li_y_Page_102.tif
d534c9d0e0f4a4767063beda5d98e6ab
f57c9146b31038d1671917f178f9a4f8b839c662
53267 F20101220_AACQQC li_y_Page_024.pro
dd9adcd483e28bf74ff92fc898352a50
b0399319af7e8a21dbf6a4a8bac9c42a3fdb704f
54745 F20101220_AACQPN li_y_Page_038.jpg
0eddadf6205b3c12163404634e87d1a4
cce730a1042d16cd5ae42e6b8a4859ecf1bffaab
F20101220_AACRTF li_y_Page_002.tif
626b901c23c91d0bf28fdfcb5495ed2e
744099efe65897d6c837f8acabc752890e136f9a
951838 F20101220_AACRSQ li_y_Page_091.jp2
30cb47cb51b9ef29b15ab79465f2699d
40c86dc4042cae6483567346d7f13b7d0bcfcb63
2758 F20101220_AACQQD li_y_Page_057thm.jpg
458d3baaf9df19a357972cdb0e9b55b4
5c796487e6b7464ca9e49aa5e0074796b946218a
1811 F20101220_AACQPO li_y_Page_115thm.jpg
be60c30c28a7b0c7b6bace6051ceb969
25450185927021b8d0e8999174f87be9a56be75f
F20101220_AACRTG li_y_Page_013.tif
343a19f304ce02aede83f27eb65f6cdb
0746f53c932848131dfb79b5981ad23c6bab1a57
913275 F20101220_AACRSR li_y_Page_095.jp2
562b82cb4d8fe031b9687e6918812ad3
790baf59c8368fdb5d19da59980fe8c95b32e5fb
34985 F20101220_AACQQE li_y_Page_055.QC.jpg
c24b4ffd2935aed76f33462234951a5b
f8e013275645e3094f536d5880b3477314bdf1c3
49416 F20101220_AACQPP li_y_Page_065.pro
48dd0c70fc706713b51220c963ba87d1
82a75b5044450a892af27f894f641251c4a8ff32
F20101220_AACRSS li_y_Page_097.jp2
4d4d7f1eb1527df9e6f71d48b38d6f20
83849aa0217bfb7a0ad03fdb679f6d3432ca571f
872 F20101220_AACQPQ li_y_Page_120.txt
d1a3dd0620bd686e2c6063a4b495027e
e759cbda15e9177ee04fbb6f0338febabe22c809
F20101220_AACRTH li_y_Page_020.tif
0b36d3dff01e4d8797428386245cc2ea
8020b5fb6489bae4a63b761dc47d52bd164b4d8c
920058 F20101220_AACRST li_y_Page_100.jp2
bb97af7c8bd50a13ed4d40de8f8cb517
f4bf143343ab2d49c2b303a79283e8b517d3d3f6
715248 F20101220_AACQQF li_y_Page_113.jp2
18b652f36dd931c3cac969d2a6e9f2da
b1e6d00285db6341abb6d75a75081aabf84cf68c
13926 F20101220_AACQPR li_y_Page_122.QC.jpg
6c557356642a5da4baa5e43effb49c09
acae60c600213ee7607bfa107a65427f43c70152
F20101220_AACRTI li_y_Page_025.tif
1aaf9327b254d889cc70b0c8558ef9fd
68ef974b411cea0d9bb14336b9d331d266947f36
1051945 F20101220_AACRSU li_y_Page_109.jp2
ba8a7afa25de93e574d79d82699745bc
6871c354fb8b01b1b618cd061feaef8d6c500112
1302 F20101220_AACQQG li_y_Page_060.txt
5b963a50ea88214c4ffb3724f5f8e23b
3aecdda377fbd4ffb7e7c447e52562bb51f57184
14711 F20101220_AACQPS li_y_Page_089.QC.jpg
3513606df25c8a56d70b988003a1b635
d1f80953a44357f5b888e7510cd38ee1b4e5fcaf
F20101220_AACRTJ li_y_Page_026.tif
c1876fd711ef81f5ea422740b3fafc0c
70068e4729e8b99717f9a7d2ba0ffc9de4c0fbf3
726392 F20101220_AACRSV li_y_Page_114.jp2
62642f76be6a7521334265d5a40fbe2b
66a1c36b41c2112b478824f806c3f8856f63506a
69448 F20101220_AACQQH li_y_Page_142.pro
54692249015fa694b59a265388567423
24cbbfc72b8ae12aed68930fcb093c9180f72297
12196 F20101220_AACQPT li_y_Page_140.pro
e24a7bec56a113cd37d0930e280a3180
7accf5b992f43af81f9bc6f076154b64eb2501ca
F20101220_AACRTK li_y_Page_031.tif
c76af4f70a7c1de57056b46bf2efb288
c58fa4d54afba257ac6f58b740c856114a58939b
814930 F20101220_AACRSW li_y_Page_119.jp2
096b8b559b449786fa2536809335e84c
c501c284f9c3ad14b8017b6f8ab3b4be6d508ce6
565617 F20101220_AACQQI li_y_Page_116.jp2
d1e1cfa404511aa85c6eaf14a3dd4fd7
71080639fee8e834e2cde076b09be95f08062a8b
1025014 F20101220_AACQPU li_y_Page_101.jp2
144b90fb47a6b9451bf76dd391cacca0
7a7f43fa59e537a706fb0c66042c76412e4bb2b9
F20101220_AACRUA li_y_Page_093.tif
162146a629768d2a599c1f71363c69f2
d52fbee7d2f8aa23e5fc596cf9e23fb5e29ae9b4
F20101220_AACRTL li_y_Page_034.tif
17bd603f8b2983cc7e4090456a4ae400
2b22e38a075b9a523bc09a9d3b0e5e146a597f11
435046 F20101220_AACRSX li_y_Page_120.jp2
2f0e771bd616d062cea59fefc70dd8d2
76b856867bf2f69b8a83d95b5ddff0d57c577296
1051946 F20101220_AACQQJ li_y_Page_005.jp2
e39b8a91ece13783bcee269e55d7ce87
ef34bd1305e30da1caa33f0512cd9833ccd125a1
1728 F20101220_AACQPV li_y_Page_003.QC.jpg
2b654bc6cd81f18021cb5a43cfecb4c3
9f9990dbc31a782153342239666c122d7d1f6cee
F20101220_AACRUB li_y_Page_094.tif
bc6bc67d6a394f235fde57422ce95c78
93ce9093658b7f5cc78d999958e839d7822413ea
F20101220_AACRTM li_y_Page_035.tif
f3f41bc91660c7bffd2cdb4c0c8fc478
23865bc62f17f1cc021afcbd0a7c5c58e84cc659
231830 F20101220_AACRSY li_y_Page_121.jp2
b84db9b7e6741548a48c1a198b634ebe
c072fa2686277ea59460e7fa8d59764f91071f26
31181 F20101220_AACQQK li_y_Page_111.jpg
889eaba4b7b257671bf2665aac2e4bf4
c2c81779fca35795f0346fb10a79b192c46586ce
57994 F20101220_AACQPW li_y_Page_074.jpg
dfa9143f6cb2a1c2ad20fee92eda9654
bd0fd635eaeebe76d0fb138decfc5f2dd3ff54db
F20101220_AACRUC li_y_Page_095.tif
0ea4e0565470a66dc221007565c0f5e4
acc2cd3c6229c099d5244c1250273d8a0127a33d
F20101220_AACRTN li_y_Page_036.tif
9c5b1b4b9cfbadd254bc292a2593443e
bea88b5806807d304ff100ecf615bd2c9c14f41a
1051931 F20101220_AACRSZ li_y_Page_134.jp2
1d25595229026a72ffdc7e18567464af
a62eb0fa84d2bb753e0c03fa6122c44933b583f9
43394 F20101220_AACQRA li_y_Page_064.jpg
0ebccfdd7f8f7bd99b43fd7779e3fa50
7797efda60e6757e8d2869707e1e49ac40eea095
50153 F20101220_AACQQL li_y_Page_055.pro
a22a0f89debc2633dd2b4f08a7021e19
a593f2f60ea69285ab945ea48e8104113c0cb09c
1720 F20101220_AACQPX li_y_Page_102.txt
851b4f5da26b155133a70018761c11c0
749de6493d0908c2becb3ca6ed68b8265abf1e41
F20101220_AACRUD li_y_Page_101.tif
e3a43dd31ceae62ccf2daba335a79b4e
ca3b165a5749f94dadf04f54fe867671589aea27
F20101220_AACRTO li_y_Page_040.tif
1fccf973d6192bb16c19feae6f25f60e
11af80bf8777fcb9cd66691ad95b0178764a038f
1392 F20101220_AACQRB li_y_Page_063.txt
3cb5afb6ca169a248778925c572758ab
8933b1257a99b1326d5d91a4b18147d4082a49e1
576570 F20101220_AACQQM li_y_Page_122.jp2
633fd48e8a06f4bc024a07655e37ca7c
ad1ab866f8fedbfdd8f5755ab9b939b378acfec0
32840 F20101220_AACQPY li_y_Page_033.QC.jpg
a65eba4801fe9da8fb20237e43edfb2c
a6c6717b47f59671beb05c1411ca379f3e937744
F20101220_AACRUE li_y_Page_105.tif
2aa9b247716fff38a08df6f640ffc96c
81c300a0a8f141f09897ea8b54b49942abe8cfa7
F20101220_AACRTP li_y_Page_050.tif
85ca4a68d103adcbbb0adac413ac55be
49518855d97be3fd683bc3c845bf5bbb778449fc
6158 F20101220_AACQRC li_y_Page_128thm.jpg
b6fa76702eeca8eb09bd89a0c7215c6d
ad3ef218d00266e8ce1481218dd8fc338cfd9373
12412 F20101220_AACQQN li_y_Page_136.pro
2763d8d30276849a602574942955485a
be492dc5e62e30d1f442334c7e7faf2e9e36a036
14282 F20101220_AACQPZ li_y_Page_071.QC.jpg
3ef1927a51031e958fe3382770868265
d016d7b6fed515ecc48821942520156f02db08c7
F20101220_AACRUF li_y_Page_108.tif
0fb56e0ee4a38a8e7053e3f4aa10140b
11da535f3b98353166823d35606ba9e51d28626d
F20101220_AACRTQ li_y_Page_052.tif
ba9d955b4e0a1192701494094c175703
8acd18514cac2217fc2cfa5178cc33172d953520
441193 F20101220_AACQRD li_y_Page_048.jp2
90d9902167d190428d7ec3ec3c21a9ec
717e26be270c1ed3d918a185da275ab36e431130
1991 F20101220_AACQQO li_y_Page_104.txt
23dd6b095bcb7a197b6745307cdc1b67
978d820346f83be0e2fbb17996e52ca4a13314f9
16706 F20101220_AACSAA li_y_Page_116.QC.jpg
2ddb3333149d5e5527798856bc166136
e0dc7a61c84f14dbfae5d577f8e95e5d728701c0
F20101220_AACRUG li_y_Page_110.tif
3e4c48812f67b042bddf0a611247b184
47eda59088550b1eea9c236c9cbde941e1fd6b03
F20101220_AACRTR li_y_Page_053.tif
5736ca3a277c68028af0829535f37e12
4511397b19ef891802e2acaab4748e97c79c51c9
398952 F20101220_AACQRE li_y_Page_088.jp2
1b4204001054c439ea6cb04a1d8ef331
711ee16875ae01d77158eaed27827d5db6b54c64
149336 F20101220_AACQQP li_y_Page_006.jpg
0547bad36a3dfba3df6574962723cf3c
df9ca4913bc01982114032558c1a561d8677513b
3285 F20101220_AACSAB li_y_Page_120thm.jpg
e0b0344d400f0a89187fb702bf48eaf2
fed90c755056cf81021377f8c74756d734dd1b5c
F20101220_AACRUH li_y_Page_111.tif
8f2b9e6422c9f0841caca66e2903099b
e4bf34f976b77763aa99fd27dc43492c36186747
F20101220_AACRTS li_y_Page_054.tif
c9143b33300c43a5697577d8c0a26889
5e3a715d6b095094fd70d112c8964743aba5e08b
550 F20101220_AACQRF li_y_Page_077.txt
1650f3c6226189892e527f0427ddebea
d737ea6e7d298edc0a65c323f57344dce1456856
20024 F20101220_AACQQQ li_y_Page_047.QC.jpg
af16b7eeaf4b3b3a3c6df20489d28274
3869035950762b10e273f188b8008fd1ae942d98
9156 F20101220_AACSAC li_y_Page_121.QC.jpg
392703730fdfb710fee4e54b16edbc53
2f2d4b6d41af4a009f939005fdbf04422911941c
F20101220_AACRTT li_y_Page_056.tif
1f19cdc130e31e29e98955a15dba9893
fe4446cad87198dda561877c48462b6e179f3cad
35334 F20101220_AACQQR li_y_Page_068.QC.jpg
6dcbffccff4730f6cb880cc1cf4008e8
c8da9c49e3afe1958cc38926422e633ecffbafe4
4529 F20101220_AACSAD li_y_Page_122thm.jpg
da0a8036fc9ea8c21ba0194790be0297
63972c926d80393783c4a9a929b5b40978df4e86
F20101220_AACRUI li_y_Page_117.tif
e20478ce6be564029a9c70094e2dfd89
c4c40f803f11274a0749cad4764bfdec7e5315af
F20101220_AACRTU li_y_Page_062.tif
30879726a0d22122ff7006ec740a99c6
5fa9c2c152cd70465bc6b164c8e26272673050d7
15277 F20101220_AACQQS li_y_Page_137.QC.jpg
ff1d883f5ff65290d0a387b57fdb5049
92247bafca17376926b151e1301477640fc5cbfd
F20101220_AACQRG li_y_Page_138.jp2
f97fd25985763107bd6de035c018f68b
beef78308a266ad145fe1c224c2ee4c14ffa76dc
19828 F20101220_AACSAE li_y_Page_124.QC.jpg
b739b9f32014cbabc93fbbca72616c51
9e8e516bf380378a0e08d2b8d0bcc9d27a310013
F20101220_AACRUJ li_y_Page_123.tif
6bc86fafb864828f68d973f40c5b0da9
fa9720503c46c2d616e2dfb3cf2c7ab879fcc744
F20101220_AACRTV li_y_Page_068.tif
29da77779373b85e898dc2da8a2f7449
6f48b5558ffb50e2f04eead67384ec31de4149ff
45106 F20101220_AACQQT li_y_Page_094.pro
ab77c46022e3cb6bf1b739e69d2285b2
a818e31d42381b6161b6210c2aed4883f0e321d8
25761 F20101220_AACQRH li_y_Page_080.pro
fbe399a67126da3f071f34baefc0fa9f
e0cf427eaec29f7246c5d7bf8cc7f79c506459cf
18728 F20101220_AACSAF li_y_Page_126.QC.jpg
2131e0107de2653ddc1509e44a4e6301
07a3c97c957f94d84a2a886405c3d7ceec775f44
F20101220_AACRUK li_y_Page_124.tif
8b07f140d378af02eea5fb81604b5503
f56c506b93e6688f4dd3b2e3e5e708f684fde005
F20101220_AACRTW li_y_Page_075.tif
fc9a5efd8576c3a867ed0b1398f7879b
7763c43c8eceaff19e02ed7196253b2b441d103c
8437 F20101220_AACQQU li_y_Page_008thm.jpg
1b491b5a0fe94789809f76041a45c234
13abd516597cee706a13375a384018af929affc7
138644 F20101220_AACQRI li_y_Page_144.jpg
cab9938b149c466a9242bb42b5d5104d
d29a53ee71e7d34c72ba0feee6daf319fd454fb4
14274 F20101220_AACSAG li_y_Page_132.QC.jpg
024cfa79a1247c2cd46050a244e1a59c
fb005d111135fe330135af5837d7a2577853ce8c
58483 F20101220_AACRVA li_y_Page_023.pro
5660a9ab98ada9cd864c107b58cd3d72
8cca9b1001c9dd6d365a0a8b3fbd20ac8c1fefbe
F20101220_AACRUL li_y_Page_128.tif
127e18c47a3e6a4c926d911fe5bc9d3f
1f1e0d496c16d2daecd920ce2f60fd39db7592cc
F20101220_AACRTX li_y_Page_079.tif
63e8b2b94c89771211268cf8c7c3f819
fce03c902077cbb3b0bcff4e28989d9ad9eea07c
1374 F20101220_AACQQV li_y_Page_057.txt
e153220dcbfa590d7e857e724d44a2e4
a60545c4d82cbdd008101bbaed6ab068bf26584b
57738 F20101220_AACQRJ li_y_Page_075.jpg
85aa9b86c3256fc7d0fea47246e38145
56dade0cdaca0a036b200d1f70ffc1d6c89ff639
4355 F20101220_AACSAH li_y_Page_133thm.jpg
4bcd654586a8cb6a629533a499053919
dcd8df94e3fc021f2f1c375c0a05b4e239117496
36450 F20101220_AACRVB li_y_Page_030.pro
51bf7b5a9a653577c0a606ff8b130441
a0846208147cfa3ca77d3e9ff1c632228e0a6c1c
F20101220_AACRUM li_y_Page_132.tif
2b8eeb003da3da3b72f11e125d2d1f91
eed2d1ae63ba052e2448b096494700c942579946
F20101220_AACRTY li_y_Page_082.tif
afd08664575aa625544062a038318c39
ca384b4adb13a36264a7c52d6fe5dc558d6d12bf
F20101220_AACQQW li_y_Page_130.tif
e18f347ab1e9111339d209c1096f1674
853199fdd2ea30134adeb5bb66ea81162ed31468
308932 F20101220_AACQRK li_y_Page_115.jp2
ca34fd86b6f232cf6ae66a63864fb03e
bd1c3b29359d7256f949fa28832ffd151e6af736
5428 F20101220_AACSAI li_y_Page_136thm.jpg
f792058635f52c0bb1134d0c78cfd66b
54d655ab82ebee2d738de691a1b46a5fa4704021
47906 F20101220_AACRVC li_y_Page_041.pro
ca355fbcbf883c0b0e9d320384cf59e8
36a1df49b2596e9ddc8a06b864cdcbb53fdd87ac
F20101220_AACRUN li_y_Page_134.tif
df608ea6da0617fdd3d4bd919c66b13f
b7321b79eef962bb8756c4b25a5dd21f25cd030e
F20101220_AACRTZ li_y_Page_083.tif
e3b85afad6e42fb1418ba933ce0ae22d
1ffe83e5dec01b9310053ac4d5082d045f239f79
4116 F20101220_AACQQX li_y_Page_040thm.jpg
af2bb522dab3e3281007a6a773ca11a9
e0fa0b3657c131d241464a7aebf41c1343cf9aff
38644 F20101220_AACQSA li_y_Page_152.QC.jpg
43ef6a431cf3b451243fc09d2de415c5
26e6a0aae34d5dcd09aefe77671787af15979bc8
F20101220_AACQRL li_y_Page_153.jp2
ee05de68867e9dfe40e48a812ce31621
00226b08be786a0e2187ab79c3b45412ae6bcdd0
8577 F20101220_AACSAJ li_y_Page_138thm.jpg
6ceca62d74289b466f3758455d2e70cc
dbf8a12fa14122889789bd0b77a999c2d26fb7b2
26113 F20101220_AACRVD li_y_Page_046.pro
363b3b546c9d635699dbf7beffc9b781
3630522c5b8840bbc2e62ca84e74a096aaf93815
F20101220_AACRUO li_y_Page_141.tif
d5df6d141c98ce17c99a73381c168f4b
1af6a9fc9e19123aaf86703bf65b70c2e55f08ed
52998 F20101220_AACQQY li_y_Page_078.jpg
3cc4cc751a877ec142e1fd2c16a9fe58
e5478f5837db12eb328b8aff1b59cb407a848af4
4980 F20101220_AACQSB li_y_Page_077thm.jpg
0df27c4c2155abed334a956692a71eec
f559a7e15f547179346c1078540ea17e04dad5e8
5878 F20101220_AACQRM li_y_Page_131thm.jpg
cbd802b983512b454adb0fbf318534d1
a16846acf706abbf12f016591e434234f5f49b30
38278 F20101220_AACSAK li_y_Page_142.QC.jpg
7aafc2328b7c9fcae435f10297c73ce6
807357345425c4fb400f220dd91d1c2a398f9eed
14741 F20101220_AACRVE li_y_Page_047.pro
b6258e29ec5b7e96f8206b5d9aa560c5
eb56681956d3d8dc00f9020eb9cd1ee79ed86dd8
F20101220_AACRUP li_y_Page_143.tif
b0648fd875f2002a3edc27fa6aa17029
6cb015ee5d19803d27f9f97c3e06739d53d10dde
16865 F20101220_AACQQZ li_y_Page_079.pro
a562ab728ac23494c960fb9f67704bed
eb1a95564ff7cec808b57c484fce62a3b477948f
2494 F20101220_AACQSC li_y_Page_141.txt
d8a815a75a3506a2bbb8a1c3794c603e
d499ebeaa79b87777341ed04afdd71a35f78a10c
F20101220_AACQRN li_y_Page_090.tif
1bf0b5bb6caf3ac210cdaba970d5f9cd
5cb586d75a486f3f345c034d6b49f2843a8c1313
8996 F20101220_AACSAL li_y_Page_142thm.jpg
c7058a1ddb75a92bbf030915be051207
72f25e624e06cce23a8dec508cfc27a7701307e4
15525 F20101220_AACRVF li_y_Page_048.pro
5227ad841380418e85adbb4f2e3d3cec
c93fa393e2f2f851013c1b8782dc9094fd5c6075
F20101220_AACRUQ li_y_Page_146.tif
78f133a215a3fa79433da4871789f085
632e2cd53dded2235f7e8d6074481e5527f2cdc6
1631 F20101220_AACQSD li_y_Page_030.txt
4d9119a78d0f333d3ec1cd6bebef8256
995f79f127bd7cb7bc6c2fa87dd4a6e0e5449150
15104 F20101220_AACQRO li_y_Page_113.QC.jpg
7d2aafe4033383f6fe9a2f5526034e04
2b9073297b588d7019f4676a06bf356d3c3fbc60
9137 F20101220_AACSAM li_y_Page_143thm.jpg
76feae97a534f9036d961ad4a624f3db
9be52b29052b88be2552b7cfbc0dfe15d0630009
32204 F20101220_AACRVG li_y_Page_057.pro
223b57ea24389b2d0d9e194824134924
7cd88a08940a21c11d69c1f1e392923b15bf03a4
F20101220_AACRUR li_y_Page_147.tif
5cd15c635b37604180b50f4f47bb1bec
9e420cd3fb27aa47b937b5a59a927af9bd683309
324081 F20101220_AACQSE li_y_Page_111.jp2
03902d623c8a83034de546677b4df06f
d8c157a9ff1fde91a3c0ae0de1d752709991dade
F20101220_AACQRP li_y_Page_019.tif
6d6622ab260fcf0dd11aa3ac9d93f659
84f8c069ae60f566bd18dc3cee412a2c80a561f0
8843 F20101220_AACSAN li_y_Page_145thm.jpg
3484182ded132852accf8af74fabc00b
084f14d3e2f454e5b3856deb69fd722d52a7ef95
51435 F20101220_AACRVH li_y_Page_062.pro
24e70c828659c35e24b3c25160ecc2a8
c7fc3667e0fd85847d11c5285d5f27ed01ca824e
F20101220_AACRUS li_y_Page_148.tif
930eb2f39d4ce12d799337f39c05d073
a02a40e5e6c0ccb0a27047369c69ef3680d052d3
5812 F20101220_AACQSF li_y_Page_124thm.jpg
a3832d30d7e8cc51c27bfe73b4f1eac2
1985aa4d619ee6ad9655987b16735bdbc42f5a9a
580 F20101220_AACQRQ li_y_Page_115.txt
33553e4aeae17a8643b48d04c8bc6134
1109864905cf54fca4b7b2ded0e3f706f24d1def
8688 F20101220_AACSAO li_y_Page_146thm.jpg
fe08a94d557f9d3b3258bb5df8a2f09d
0526777dab6ef279b98b8e7dbc5ca905d88aade8
32858 F20101220_AACRVI li_y_Page_063.pro
6950c16ea6589cf2ae2679cac0d5977a
9cd3a27b2121bb9a372f09b0a40870ed830470e5
F20101220_AACRUT li_y_Page_150.tif
ffde63fd521519c4d92cbd7496739cd9
af68b0dc466ae0b3308742fc4cc7032487e39774
F20101220_AACQSG li_y_Page_042.tif
047d52f85d1439b1bf611f254a33d232
3c9f41ea3530aa7c9447cbd52458fdb75101ccfe
99940 F20101220_AACQRR li_y_Page_036.jpg
ad8b0be491079fda87ef7a5c45cc0834
b17d9464dab232ed660c6c149659b6b4fa7ce1e2
8497 F20101220_AACSAP li_y_Page_149thm.jpg
2eaff91579afe99db1b93a9ea810d5c3
f651bc9b6009e07fd4d6e47c9fd9159cb088a038
F20101220_AACRUU li_y_Page_151.tif
2155a08fd24ac9c6f17e7cb113cbe5aa
effd959a0ac4d65ad3516bf302e94222e7575292
32200 F20101220_AACQRS li_y_Page_093.QC.jpg
9340b1c12a9a16005e3e7e8c7b210e62
7dfabef057a0d0da758960b4fbdaf8d14f5ce9dd
9002 F20101220_AACSAQ li_y_Page_150thm.jpg
566c6ddd0a5e7598b0081c46781c3c29
30579199fc68523cf8ca8a5129c5b33fd3a9cbaa
53479 F20101220_AACRVJ li_y_Page_068.pro
748dfa6d28ad6adef5023be58f33a6f3
2b74f3faab8cf67cec4dfdfd5688affa50b7c7e4
F20101220_AACRUV li_y_Page_153.tif
d1b28fa333afe4a2f7bbb74ca42d455c
a9176f55f5efb97ace0475a2fdb7d08aaaa2101d
F20101220_AACQSH li_y_Page_119.tif
b1c29a157a5ebabe3bc0cdb60461f26d
9de1a0d9ca0e37c3a96d40d8003f72092133511a
3509 F20101220_AACQRT li_y_Page_002.jpg
7b666d1fc881afb83c497dbecac16677
c7f8cd7042acd086f68dd49ee8a39216a349c485
175815 F20101220_AACSAR UFE0022512_00001.mets
33d74cd928911062c3815715b0b3146a
a6550941c7665afc54647ab92caac34529572a9d
52559 F20101220_AACRVK li_y_Page_070.pro
7216cb9a202aa33c5b29fa079c5f467f
984d836e70dd12eb200a22d07cdca3a15437435f
696 F20101220_AACRUW li_y_Page_002.pro
68bbfca6bf99109fd1c110304b06d0e7
a8300e7d47f658838bf46da63f693a19d3e70cc2
60155 F20101220_AACQSI li_y_Page_047.jpg
c332c24aa7f89a07eef6c534f48cd560
1ac9209472076cc8610605fa22640da3757e0cff
102654 F20101220_AACQRU li_y_Page_104.jpg
f9399e0c179c8ec0f6ed98f488b75e86
8b782af6b7debabf41e7662ac70ffb64f9cbd2c6
24923 F20101220_AACRWA li_y_Page_129.pro
7f1fda0708cc17d7b51c53f73948896a
3145e888e446b9b465805480d8112b7f38761b18
39107 F20101220_AACRVL li_y_Page_075.pro
307f7999f0203ebb5a3ef50e5e65461f
42b0b840b2253855849011a7af8f6b9c25f43c67
44287 F20101220_AACRUX li_y_Page_013.pro
6d3c8889fc01687c4bab6aeb47d24f83
0ec482df0efe2404fe23f676a21a7af0615185e4
F20101220_AACQSJ li_y_Page_008.jp2
e2574fb47eedb3c1ccf494f333009612
a8d27da6269ab42388a31a50018d784a150c7de2
2187 F20101220_AACQRV li_y_Page_021.txt
acab564591cec6a2fb26531171ee2f61
a664c6a33e21d0960245d47d0d77cfb9f75e54e4
13696 F20101220_AACRWB li_y_Page_130.pro
395207efadbf6f48393f3234521b07b6
ec6dbcc612226b8c1348e7a47fa6cdd810114152
41011 F20101220_AACRVM li_y_Page_082.pro
881df94d891fe22108f9a4b3b44e5095
392f505cba36a42a909fc8b85872bc93e0fbce48
54903 F20101220_AACRUY li_y_Page_021.pro
ae4306950cbcac2d3291d050a47afdc5
bd6ecbd30815c80556c031fdaa77f0ed2fce7a7c
23384 F20101220_AACQSK li_y_Page_046.QC.jpg
0e2c62e18a452d42ad8512cc80bbc5ee
d842d656b72d204a5d73d9dc2e89eeee7a70047c
F20101220_AACQRW li_y_Page_103thm.jpg
f97944986873c2f7126d24db396ac361
bbda7da33b90135d38d2a4d51d74a0e11aa1ad61
24054 F20101220_AACRWC li_y_Page_131.pro
5b4a687a16542c7c6a46312d6663e613
06c80cdbf7b88ba38957d764c7c1bbe23ef18f26
46217 F20101220_AACRVN li_y_Page_101.pro
5540c50d8b40cc410b07e3130d52a047
2554d501a30d1f551eaabe8e9986f57aa79c2912
54321 F20101220_AACRUZ li_y_Page_022.pro
7414def5815159e2dceca0109094e53b
9eb7a63848253fc631802c4c8f2acdeca26d4b29
98828 F20101220_AACQTA li_y_Page_054.jpg
0391eae3b37ddd5a57f8e7340b52f70b
bd8afb8c30bc4424df1eb7ea8dd9aabd168edd50
5249 F20101220_AACQSL li_y_Page_125thm.jpg
db6939f0c5bbfef6ba8fa32fa45f2fda
455712ebd9dc38a66919d9582ee467ca34a94fd6
8541 F20101220_AACQRX li_y_Page_154.pro
18c341860c317440c1605c96dd83109e
94b2ba7bb858148e2f4599d39f0df60c84335ab9
23188 F20101220_AACRWD li_y_Page_135.pro
5c1cc3bfc6c484f06352919f9f85a808
40f29a38f47e4bc395d43d66d602b6639906c291
42057 F20101220_AACRVO li_y_Page_102.pro
e20c1e97682cb304df1f1c0aa7dd4cfa
b11b9d2a45887055f9a61acca6f3ecd3d23f64d2
F20101220_AACQTB li_y_Page_011.tif
5fefcb0eb0c02cdd4f8c9e0fa22988e2
321f22e5b8d19b39ef43dd266f1a5839b5de8cf3
3675 F20101220_AACQSM li_y_Page_114thm.jpg
76daa24b96c3a10da358b1eff7eab175
2376c42ee95423686ec8bfa446186123dcdcbf66
1878 F20101220_AACQRY li_y_Page_110.txt
ffbb094bb47a3c3fb952a3abd6ff696d
d4ed8b32aad6ce3ccda5392ba9d924fe08737ec5
11416 F20101220_AACRWE li_y_Page_137.pro
8561dd196830655c36bfc889b6aa5264
570fafe3bf2205e1fc5b1db0a6bc985b28ab859d
46453 F20101220_AACRVP li_y_Page_103.pro
932b29864a387a261f74a251c9277eea
629290baa441205ac0db905c6e99a313a3e2137b
27743 F20101220_AACQTC li_y_Page_140.jpg
127965e2ec2a0af0081c8428325f5c7f
dafed0c816012862abeba2194cd6949b3bfbfe3c
567 F20101220_AACQSN li_y_Page_111.txt
2060b8a1ffafb10a7c1d12b21670d417
f49568498534ec0991bea56f5463434ca4d493ca
F20101220_AACQRZ li_y_Page_097.tif
2f7798bd2377af40c10f5e908897023b
6f5b0fe7d2b70afcb69fe7b8f854ac14b0d98a76
52818 F20101220_AACRWF li_y_Page_139.pro
8f2759e99642c7b021e916f9d51990a3
9ed1753f5183669c7b5b1f00bd5dbc772800fb39
52012 F20101220_AACRVQ li_y_Page_105.pro
3d8dc693772457d0412e3be04f16efeb
629898df1babcfce433a6dc9aabee5c32d0a193c
1051934 F20101220_AACQTD li_y_Page_041.jp2
b81bbed9286468b21ab06901f2c4baa6
e5788fea9984dfbf864759b26315cb1b2484c9e5
1582 F20101220_AACQSO li_y_Page_037.txt
c75aee815d7dc511dbdc6705ba846412
362191c921b2cc1f3409213a766a28a53fb75493
412 F20101220_AACRWG li_y_Page_001.txt
4eb28d0c8c3f96b16c7c7403438349e5
c8009d2d1aabd51ca7dff3b068300b7258fec8a8
44463 F20101220_AACRVR li_y_Page_106.pro
8a9b8e1f17f59dd3d92ca82d5463108f
976fb30c0e56358ecc663035b3de84a9cf136867
925922 F20101220_AACQTE li_y_Page_125.jp2
a64250c5b208199500d0873eb5d9ce21
fbcc83f471ecb257f830609d981bbc87b7aff90d
7626 F20101220_AACQSP li_y_Page_092thm.jpg
6e10b5b44cbfa1162c8c9a963f872945
c68cb32a64908cc12a75da1fc24f01240ca04834
132 F20101220_AACRWH li_y_Page_003.txt
1fd67862a001f0381686c4f31896366d
1127ef4203c16342bae820233a06ce98a5e4ae76
48926 F20101220_AACRVS li_y_Page_109.pro
82c4b8dcc8cdc4bac36c6cc4c545267f
6308146d7f787a8ac5c6f51497521cbe178c526a
22773 F20101220_AACQTF li_y_Page_079.QC.jpg
20aa026c1fbf5eda121dbd992c4977b9
f28ceba57b1349eb016aefa4d54cf1b4f5dfdb45
897788 F20101220_AACQSQ li_y_Page_128.jp2
99444ba69f13fd65d83f3954faea29a5
52240b210b0daabc96af4320cfb23288f0b8235a
3674 F20101220_AACRWI li_y_Page_006.txt
3592098fb5a51c9da6dac606a3f4c155
76a82ca4810e721970310acaad2da14a34d68db0
13994 F20101220_AACRVT li_y_Page_111.pro
2556a9c2a5c9a74362bac8f323116c0e
af4fb928e3aa43ee1e48ed84e48a5671214c408d
42122 F20101220_AACQTG li_y_Page_092.pro
1c20829a4a655a06b3cb94a87267be4d
98c0ee58c27d0f51644f8eb1138f87e696ae06f5
4494 F20101220_AACQSR li_y_Page_132thm.jpg
50a9c05c1867bedfe02858d7361d74fd
edd32580f80136b01666303d43f4f322c60f6eaa
2406 F20101220_AACRWJ li_y_Page_008.txt
357cca086f8e523fe77e51ec1725373a
71a73213885e0148829a3dbbb8ec6d0655bd8eae
6437 F20101220_AACRVU li_y_Page_112.pro
2e0e9e2ddcf34b43620ef0e30d614e1c
b7fb70bf5a112f16654da99de46d2cf55c3c5a43
496228 F20101220_AACQTH li_y_Page_027.jp2
e16e20d10c7e46330f2bbbcc504809fe
1aa7cef5b06c45828a11770254aeeaedad601b55
51058 F20101220_AACQSS li_y_Page_067.pro
de2af88f861b419b7a3cc8896afa4ecd
e6e45e34a701ead18579b8dba65796fafdd134a1
33482 F20101220_AACRVV li_y_Page_113.pro
8f44a3c8e467fabcc3a2a6ead2f1a46b
763827738925dda493fcd2f1e4f079081ba784f3
18807 F20101220_AACQST li_y_Page_131.QC.jpg
b5f225fd3026dc203649831d3394d784
5c9886770041ed8f9b735e28ce2d6adabb36167e
1963 F20101220_AACRWK li_y_Page_016.txt
2cf4da643eb382ecca676486eb5a71ff
1c9a2e069535295a15c0dfda27c8885b376278ca
32072 F20101220_AACRVW li_y_Page_114.pro
a8ff18713ddf92fb7f47aa2fe2a456df
3f34d4e35a3435e104c6f0fe9e626d422a5fc865
F20101220_AACQTI li_y_Page_149.tif
2c3e619d68594658fee8c77372e083e2
8485681496dfe6e7098ac3db46be59d15b10667c
67747 F20101220_AACQSU li_y_Page_040.jpg
55b3085760d2720f46960069ad5cad2d
28b59399c733bf1e7b642d817b54720884180b84
2072 F20101220_AACRXA li_y_Page_070.txt
518f7dad33571856baff2e32c30ce378
3c04d2dcaf8f0097a2174866724050aa77ed05f7
2054 F20101220_AACRWL li_y_Page_017.txt
dc7b44d8dbd7ea3f773fdc18fc58ec7c
fcc016be767a5f3416a41549c92fa4d855e3253c
3214 F20101220_AACRVX li_y_Page_123.pro
1e5845da7681c57ac0c6b021321be8b9
7186f3e5e7a68bce3b0edb1a7ab9c0609d75e520
7488 F20101220_AACQTJ li_y_Page_106thm.jpg
c90317d6b067092248fca80dbe7a7e1c
2694386eb226bb1d9270623a61eb27909c972f79
8301 F20101220_AACQSV li_y_Page_065thm.jpg
45448d7f3cd4a75c98a9d5c7ec9f7914
01da149ac918b8b486623eaa8fbeb61ae7f70c92
1576 F20101220_AACRXB li_y_Page_075.txt
072b6dfabe2ee4a47a0480d1636cf20c
611782e6ba17aad9ec0779b51348c4f84d02abc2
2129 F20101220_AACRWM li_y_Page_019.txt
74f31db239337de07f96d4e1a9b174a7
664e98d3d35e089edf0d2c8da98a4d8520672502
22887 F20101220_AACRVY li_y_Page_125.pro
57c421a5448f6490aa9271b7ecfea23f
9ccae987151d3db059a840fd552119e738d69fa0
35142 F20101220_AACQTK li_y_Page_139.QC.jpg
e869138cff197636232b851a191efb89
72933c9ed6a13b3d04457aa6b79c83b48c5eec8b
8110 F20101220_AACQSW li_y_Page_051thm.jpg
25ae795f4603e9a0f9e64ccf96221707
a90970fa30661ccf88d7aec08f82cdad979fdcc9
1666 F20101220_AACRXC li_y_Page_080.txt
729682dfe185bad7322e6d18ccde6f91
0109764da09eec8e7e2fc17767bc87152007d4d3
339 F20101220_AACRWN li_y_Page_028.txt
337cc0f63d6de2bb2fd2eb1e08361ecf
8cbf47b5bddae7b1c6e9a002df4327c6bf288534
23512 F20101220_AACRVZ li_y_Page_127.pro
cd9a6d352c7ffc3671ccd6168590dd01
b97e7bf3565e30479f438bfa799d5258780d4599
46403 F20101220_AACQTL li_y_Page_034.pro
c1afa92ecc5193629f72df0faf6b48a2
82ecee8b565adebca34f4074a7f1069f49a1d61e
8008 F20101220_AACQSX li_y_Page_101thm.jpg
32025e8297adfdb1f338146f74b451a2
aa6970a6860f90d1b1c06e5100486670bb2ca565
69569 F20101220_AACQUA li_y_Page_063.jpg
e1deab59ce5be1f53c32cf0d21058ab5
07a3855c514704593bf629f5a8fba775e9802afa
1649 F20101220_AACRXD li_y_Page_082.txt
96e32977419bf89687864ae80f9fa23d
cc55846db3c8213a8c0d7dd3037f06ea290b6b11
F20101220_AACRWO li_y_Page_031.txt
f0a57ffbd6f65309ea9516a57965dcc0
c62ef4cb073adeb319501945fc8a1e2ab9e39346
5298 F20101220_AACQTM li_y_Page_137thm.jpg
c85661b3af50f27e2afe0d5bb1cd7ee3
8316190af01aa4e7d5e76b2c42970c837eb17c70
783346 F20101220_AACQSY li_y_Page_047.jp2
c33c5eacc2ed55e3c773036e7529fc2a
ace1e8456c101894b64807e11bc14a98cc0f39e2
30966 F20101220_AACQUB li_y_Page_076.pro
60316ff757fad2217f5de6df4b833ce7
9b899f7df11443c9eed7709f889d60e5b855b0bb
1758 F20101220_AACRXE li_y_Page_085.txt
148ce43bb88d3bf6b653916ffbff7df3
5844fd11e56f3849efcec04d722ba2493216d8ba
1864 F20101220_AACRWP li_y_Page_033.txt
671b21486981e0dbcaddf9602f99005a
62894c807503ddc88a6257ef5424a1dd63073a75
22150 F20101220_AACQTN li_y_Page_038.pro
a4317301be10cead81a7193bef835316
00bb8623ff48ff2afe156c088cf929b3c1191e4b
F20101220_AACQSZ li_y_Page_084.tif
23b2c3a319d052f89be59ee50f1ca468
6d128079f5b6d32649ab9067a856a0dd4b291afc
7355 F20101220_AACQUC li_y_Page_081thm.jpg
653107f21824f75c4c0c57164db07415
88fab82e2ab05d93a05ae3815e6835949e732fe9
780 F20101220_AACRXF li_y_Page_089.txt
0f166ede8cefc0793094064e392cc1e1
b61eb391d7b621a72f033cc91990393a87ee43a6
1874 F20101220_AACRWQ li_y_Page_034.txt
7425e7e892ef2c3e58fc550439084fd4
392726b23ede2c6135f0adfe2bd14d57df1f380e
F20101220_AACQTO li_y_Page_042.jp2
b8911ac4c034fd37451bd3aa79f82ed5
6a98172919ccc98a2138bfc24dabb6046cbfda3f
29483 F20101220_AACQUD li_y_Page_102.QC.jpg
60670401af7f3b86a9ac64b7904ed509
54e7aebfba2fcb8f80511f68ac80eb9fca9e0f36
1876 F20101220_AACRXG li_y_Page_093.txt
b7fb55c1e228bd1183ba325be47b9ad3
db3a4644a8e7acede44ac9edaed99670a1dfbb93
1490 F20101220_AACRWR li_y_Page_043.txt
8ec761ec458bc9660c2396db3d798eec
d3113ffa7ad71f7323bf4e536851ce9b5003b175
10752 F20101220_AACQTP li_y_Page_111.QC.jpg
48507e6386c5a92aac1912188b1b7467
fdf3afcd383397da3a10f18cdf9cc0140a7f87c5
F20101220_AACQUE li_y_Page_063.tif
663df042c54b0c76076629109ab1b9cd
423a73426c8d73cc18eb8024b77b9234ef2cd83b
1994 F20101220_AACRXH li_y_Page_096.txt
19740ca048f2acecdacfffbd90044730
a3d0358947bbebef1c7a7560012bb7fa963f8368
1138 F20101220_AACRWS li_y_Page_044.txt
6d18bdc20ccb88b9790a4ed58bf53e1c
b80e169c7249fe507ddd1234edc316587c3ce385
19439 F20101220_AACQTQ li_y_Page_127.QC.jpg
41a1040f3bde5ed7eedb2f23c4c38011
5ace94dbcfc588137d206d05cd76f83b793af020
35618 F20101220_AACQUF li_y_Page_146.QC.jpg
b5df702efcaa8274c4469cefaab770a7
86c45d9129ce500aaec61770d986a82f831cabac
1693 F20101220_AACRXI li_y_Page_100.txt
d06c40c07b9c68414c8794d128ae57a2
e1b24dc8a2a8aaf3ea93c7eea2083bb5ac0d309e
962 F20101220_AACRWT li_y_Page_047.txt
46af993dea9868e2f7649c65e1f42a30
761464fc27a3323a25ce7ec47c2a4215d649500c
103680 F20101220_AACQTR li_y_Page_032.jpg
6ff31b60454c2856bd44a2b5cc1a6584
c06327194f3ef03337acf6b9b4007e8b854dc125
1051956 F20101220_AACRAA li_y_Page_107.jp2
5de9c5746e35febfc5068c995678d2bc
7b33431b1fdaf90d4105f5810062753b1df047e3
908 F20101220_AACQUG li_y_Page_027.txt
9d993fe50a86166035a14a1ae479df4d
116d07661e2c916f65580313dcbd9a29c191bec9
2049 F20101220_AACRXJ li_y_Page_105.txt
2872a0ee43a1b76af861c80f25fe9dfd
d25f8f8e255b868ef72cdc9f3b326c67c654d992
592 F20101220_AACRWU li_y_Page_049.txt
adbeab54c25c3a146259d5dd0f6708da
e37461eb9be07184303c8aab65da15de9e88a214
2006 F20101220_AACQTS li_y_Page_050.txt
ee7bdee50c0a93cacd50b6fbd1754286
6cc9b4241914a02c628b14cecd2a56da4e978fa6
34023 F20101220_AACRAB li_y_Page_057.jpg
91e8e291d8b65878176c5e0d29071efa
05a0173172ce6b37ad90cbf09574aa0d0470a2a1
39412 F20101220_AACQUH li_y_Page_098.pro
0e1266d828ca58f748509412c4d7367e
f1e608e5c1f2f6021fbec051b7f87e557961549b
1981 F20101220_AACRXK li_y_Page_108.txt
97b05c84b0656ad53791f8959f44f269
1b1d57d37d44ea07b2885af6d69163bd8ec454bf
F20101220_AACRWV li_y_Page_054.txt
87a58c11558e47be4aefb0f17bf7e575
29810c293871c4d3dafcd3caea069b1749ba7204
725 F20101220_AACQTT li_y_Page_003thm.jpg
c8549ecce7a10a8581e7898301622079
85732844890dd7778b1c1e09f68c6cca89b91c4c
138194 F20101220_AACRAC li_y_Page_009.jpg
0ba1a397dd7e48ceb58bf4efffc18aa6
335a6359b08f12e823c8e89c01746c5b66314be6
8151 F20101220_AACQUI li_y_Page_054thm.jpg
de45ff1b3dd7144a71020185481e24c4
1df016bbc43561e626c42fd3b0e77e34c7df0b80
1132 F20101220_AACRWW li_y_Page_064.txt
1b9351227849a9397027650ff154924a
7957414ea1515dfbd4ee22126641a0cb2b6c04b7
2078 F20101220_AACQTU li_y_Page_067.txt
7c4ab1244f7c1b8fcea6bbdd397445e1
6d987638b446ee99a868838e69e1e8dfa13d9f61
F20101220_AACRAD li_y_Page_018.tif
b9667bcf78bd77ae0b910177f11a25d1
d780f696875a952df457a6022ac1f03bda6ef2b9
8511 F20101220_AACRYA li_y_Page_015thm.jpg
ff62a45aac29338600290778f256714e
abe5c7e1e783a318e430b4f409afb0dbb0a83637
538 F20101220_AACRXL li_y_Page_121.txt
4de491e535671c8314da276a8dfc3f4e
f82c5a16fcf66e117eebc32594d7b831ad241b85
1955 F20101220_AACRWX li_y_Page_065.txt
99eb21a79b0225485df27cee4895be8f
0f53c9cf883888c472b28377977667ed9c12c668
133601 F20101220_AACQTV li_y_Page_147.jpg
0baeb3a72bb5af6f283a7284c442d94a
7c0f3b4fed7d4efad0ab91ed80bd16fe3a045007
47298 F20101220_AACRAE li_y_Page_084.pro
a13914f1b9115f03949bb4c5ff0508e5
033d9701d05659809e3a9f8eb372a32bed39c78d
F20101220_AACQUJ li_y_Page_103.jp2
af4f4a23c69c4ff29c5d6f3370a46f57
33206a64b4b2d0a6a61c2f37a52bca728ebb62f0
8542 F20101220_AACRYB li_y_Page_017thm.jpg
67778a81b41ca51f8a6b8cf2b9034d72
6b784518dc7b76ec0b7f90a50116267ac6af845c
301 F20101220_AACRXM li_y_Page_123.txt
6c522172df0ada30c761463f82ba91bc
816f5515542d09178c89063566cfd6b1d99eb275
2114 F20101220_AACRWY li_y_Page_068.txt
020854acc7aa435fd648d785b76416bb
8a2ad28b2fc4d24e75a0bae123ab2e9553fda6d6
24701 F20101220_AACQTW li_y_Page_080.QC.jpg
922ac74f813fbcb71080ac1b3c8ed581
bd34f887c0233b0531706426a97f0c98dc25e57a
37397 F20101220_AACRAF li_y_Page_073.pro
15924b254429ca43de8dd0908d45d150
21c3d07b86509740576f9916d381ab74e878da5b
67544 F20101220_AACQUK li_y_Page_014.jpg
e89bfce2e99fde99b59dccab676e1fdc
6d80300907d60b1ec0f42daf184ec8f007b625f7
35623 F20101220_AACRYC li_y_Page_022.QC.jpg
62e9383b8c2a59a5c018c2e9655bcca0
9025c34bf4acb23406375488837beca684d3314a
730 F20101220_AACRXN li_y_Page_124.txt
53b7d9071e9ce96bb9017d96b0290b37
d6d5c2c72b14b1a6c58cd69bc5e0cd30b3a36e7a
2160 F20101220_AACRWZ li_y_Page_069.txt
2823469aab3ccdde87c7482d09ebf093
d409476ad685453fa3f6d2f9778c89ce997eaca0
13667 F20101220_AACQTX li_y_Page_133.QC.jpg
0405dac6043f467b2215fc8c6f4934e6
6d133c545686574a6126b55b9e8e361e316e45ef
F20101220_AACRAG li_y_Page_118.tif
2c24b47136fe5f77ffba4025c4b30e4c
6c43ba030c4830160606c9d8bc62fa136a106004
6262 F20101220_AACQVA li_y_Page_003.jpg
9c3eff541ba5150c1488b4acf238b0f9
0d64347bad605ef25285e619adade04839209ddf
18814 F20101220_AACQUL li_y_Page_090.pro
b981e358fdd38e1a5e031df65b0ee244
37050929048c5885b06c3e86e35957a61df60f99
35923 F20101220_AACRYD li_y_Page_024.QC.jpg
d5cd5028edd482d8ffe4bacfd1ad33ee
ff6bc0e182ac0837e76536d4c39fedf8226a5b9f
740 F20101220_AACRXO li_y_Page_126.txt
682ca0a361f08478cadf5e6ba50bbf97
39ff4ece1653dee22c6933df2e755bfdeb716bda
2101 F20101220_AACQTY li_y_Page_061.txt
ffaf02536eb34f576a64119c133aa811
795e11a657de44cc6037ccf87d0c61576b374563
38262 F20101220_AACRAH li_y_Page_151.QC.jpg
d126e0cf48d73eac2a44d5acabdce4d3
4f30a16c43f8ecfcdcb3cc9cdb7683087cfa8d77
42590 F20101220_AACQVB li_y_Page_091.pro
8c69725ea887c88002cefa1349ce84e9
ba8a9add7ff7958250dd71ef06c4373758fce149
7900 F20101220_AACQUM li_y_Page_006thm.jpg
289122f221cdaac2dc5c4ed5b1ea952e
ca370d7bc68a89eb63e1a623cdd2d40a27b09679
19093 F20101220_AACRYE li_y_Page_026.QC.jpg
dec0bf3107643ca25ed49cff05bba684
a33198b800e39440f9ae19b2cb1ba8d5b61decc6
1187 F20101220_AACRXP li_y_Page_127.txt
251e3112f862e835a705c4a33380322c
53597865117622af4720f3b8664037f8be0e514d
89716 F20101220_AACQTZ li_y_Page_102.jpg
85331434b8df703246f84bc672527b79
41f3a05cc54cbf0f6659d088676257fb31a76b2e
F20101220_AACRAI li_y_Page_122.tif
319043ff579884b22545389b976e5c44
e74093e0b509a824884bc5f76f75b862ec4a8f27
120265 F20101220_AACQVC li_y_Page_141.jpg
d9c92c8ee4e0cbbfb8e197c8fbe23570
0ba6ab26c8254cdd14315ab75818bc3fb7bc4141
35123 F20101220_AACQUN li_y_Page_010.QC.jpg
2574ade43d11892d9168c7cd820583c0
c42240586cb5eee788c0db9ff3c88627415f2f15
11536 F20101220_AACRYF li_y_Page_028.QC.jpg
4fdec451d9baec550ce7ab221490ff8c
7ebca521f52aa2e9e07e7dcfd7fddfe24b44e02b
794 F20101220_AACRXQ li_y_Page_134.txt
98511c54ae994c0b745afbbcb81bc586
5b15bf69215e88c69975412e61becb679e920a08
27882 F20101220_AACRAJ li_y_Page_095.QC.jpg
c35342deb4c838ab156afede6a514893
4b6c0f1953dc6de7dddc58036fb796d539f28185
3899 F20101220_AACQVD li_y_Page_007.txt
02a7d05c44e3727fbe7295bd71b19cc7
ea5a4da85e9f03b540f2f4f31dbea4e4b7fc9788
854640 F20101220_AACQUO li_y_Page_124.jp2
3c4ba83870ec59a003966951369938e8
d0effcf442f340bf6f6d9442ab20c30cb29b900f
10664 F20101220_AACRYG li_y_Page_029.QC.jpg
cb87b23910c51b07b0842077d1d701df
4db81b7473be84901c51868f900e5dc82228ef85
1089 F20101220_AACRXR li_y_Page_137.txt
7b566ce057b45761600305fbe23945f5
52fd7958e09c81fd3bf56432fc43f7f76de66886
48481 F20101220_AACRAK li_y_Page_050.pro
0a5f891c87c299c4e7bf4b8f10c11405
baf2293369be08e78a4e6110d31a8571518d6e7d
81190 F20101220_AACQVE li_y_Page_030.jpg
4d9fa239eaa2879636e4d3a3cf67771c
277732af2b878412378443c1e90976c10e2275df
945067 F20101220_AACQUP li_y_Page_129.jp2
a9870766df5280ba8daafacf49884c93
9730baee6f8ef3a88992ff5730ee576b361cda79
8033 F20101220_AACRYH li_y_Page_033thm.jpg
f1723c811793322c98285c9958bfbd49
2a3bc76e9f3483cb61ddd135a8f3a4e76ccc3d62
2639 F20101220_AACRXS li_y_Page_146.txt
76bd388da42555f1b203a5089c50464d
aa33beec44b907fbaae7cea4a92dd43e2a3cfd5d
671467 F20101220_AACRAL li_y_Page_045.jp2
1da60a5264cf5d1ebd86ba3feb69813a
aec1a2d46608829bc2709dae510ad589671b9bd7
138223 F20101220_AACQVF li_y_Page_151.jpg
9e921c3fbf669b7b83ef3b54ecaf54f1
2af63e0a1d4dba58dcfcb76e8bf2b7fd46d1e71e
F20101220_AACQUQ li_y_Page_099.tif
0903943875f072d417f1014895e2c903
22ee8e30f80fac3f65fde3c94eb128b6893f4af2
8087 F20101220_AACRYI li_y_Page_034thm.jpg
97e9140101511862c2f698ebcbbf7192
acc8f13b29eb86acbba262b7fe1e925da88e6ba1
2697 F20101220_AACRXT li_y_Page_150.txt
8fb5cbba857406e271b6dbbc5fce0812
689b45e2946c900c688775af457a5a7333d45cb0
F20101220_AACRAM li_y_Page_133.tif
c6c839df02e2ea36dcd10645662c576a
ba159f6c9baa25dceb50f2229a1b9938d87cc687
68923 F20101220_AACQVG li_y_Page_153.pro
c69c6bd1ba245b533ea77dcc69bf45a1
07ce0423c27dc4e62a10e65a5884691cf1b66367
629328 F20101220_AACQUR li_y_Page_056.jp2
729bda948d4a84b0cc7a9e28213f3425
7012f5026af66e337a18a68256a812f2d37deb82
8908 F20101220_AACRBA li_y_Page_024thm.jpg
51b338dca56a68bda49c97c8cecd4961
38e94ca2a59fe4ec48e7557b4e91b7ea18de0780
34247 F20101220_AACRYJ li_y_Page_035.QC.jpg
bcac4176e2468ee43e4ef6e679f6eef4
aca69daa5aaf1fe3d427d6e78ab2ac517d1ea823
1960 F20101220_AACRXU li_y_Page_001thm.jpg
8b14b17d86940fe85fcb5c0fbc555c27
9c6ea74c0974ac3c3e80725d6bf20ab8f60af153
1137 F20101220_AACRAN li_y_Page_133.txt
cb9ec989b6f32d7120b8f66c3af911cc
cdb870f6c567e09f494afc9722a420ffa7d88f9c
1068 F20101220_AACQVH li_y_Page_002.QC.jpg
5b2c2b6a0651b511e2c224da6c6c9b87
6968cc606ccbcf445e738382a4d6cd483b0b0750
8637 F20101220_AACQUS li_y_Page_022thm.jpg
5df496184b8c8f699e91e805e88cb3bc
fb025f8fd6872ad98ff32f692afbfaaa8549da28
2234 F20101220_AACRBB li_y_Page_020.txt
ed3d5702d2d4e03eaa4a64cac3787931
d5d4482b5ccc9fd6d8016dd4ca863b7c6c863a0c
8201 F20101220_AACRYK li_y_Page_036thm.jpg
2e8fc89c6977d7db3ecd9fe413254623
bd7b921aea28ee5ac228c262754c2ef6d74d599a
478 F20101220_AACRXV li_y_Page_002thm.jpg
97608d0d73fef5c5c44b56dc4059a2a8
16913757db488e8bf768a169e3c7b8e15eb944fb
1702 F20101220_AACRAO li_y_Page_092.txt
ef7580301fb5de8fbfc95eb9cb74d687
61064700759b4274203dc5c38e2d765c69dc0e2e
4811 F20101220_AACQVI li_y_Page_042thm.jpg
dc3a8af3ec365869768f6806b2e3a3a9
a03af84200f17f515e078c20811aed92c70704f0
8839 F20101220_AACQUT li_y_Page_068thm.jpg
2ac5de3ca2ad006cdcb8ae5049895000
33f832fe6ebc6352e4f3617321ca62f105cfba0e
4614 F20101220_AACRBC li_y_Page_117thm.jpg
19b0b2aac257cc1423395992d09be073
b2fd49e14095170f43713e9b2c89a29a2d4c694b
15533 F20101220_AACRYL li_y_Page_038.QC.jpg
a05e5d3221f70ee998212f92a75d2466
05c9dfa696e06785d0eb27884789acc8bd95fa5d
6655 F20101220_AACRXW li_y_Page_005thm.jpg
39ec4361ec6254c47ed2b692fbae8b56
2b770ea038061c8b060dae77c354bd723f774878
32691 F20101220_AACRAP li_y_Page_029.jpg
2ada9a9d79ab617c7a19ee2c94facc6c
7bc9134e0831d446e752e6fba51311e85c2ae2a3
1051981 F20101220_AACQVJ li_y_Page_145.jp2
afde6663d0ff528544211240c0e72a97
92d2a4e7ebe22787e4c53e59f43cd51d2d40108c
F20101220_AACQUU li_y_Page_061.jp2
02556c1fd248763b7d62901436c18093
5571539580a3e0fcb292da002492d808d3c6043b
7334 F20101220_AACRBD li_y_Page_001.pro
e5fe820de585c9fad9e46fb9f35d7efd
ac337869b2538b238c8107ed5bf1977ba69ff2ad
7575 F20101220_AACRZA li_y_Page_060thm.jpg
6d25afbbe4c3d1b3c3b90b6ce9d89c4e
b46b07e44a54b40e310461558e850a36921b1daf
33301 F20101220_AACRXX li_y_Page_007.QC.jpg
d76b94f56931f2547956132ee2a7a63f
3ec6da895068e67a9116d1998be5c2048a545cc6
57728 F20101220_AACRAQ li_y_Page_072.jpg
85f17b5fcd0396fd1f48e65c049c7cb2
e70ba8064e5f47ee0fd71fc5433202b47c315a41
113818 F20101220_AACQUV li_y_Page_018.jpg
74998b8e0e1be032d2cc8b022e3466be
12470e6e6dcd7bfa07078c9e974347cbc06537ab
107532 F20101220_AACRBE li_y_Page_024.jpg
96ba11e145c75ab957c44fc6f48286dc
be28acce3eebc0183115268fb269fed29da40bc5
29050 F20101220_AACRZB li_y_Page_061.QC.jpg
adf2b62324526f630e07b84ed7e0489e
67aebf3991326df05e69b2a10fe823db3bea3c5e
13836 F20101220_AACRYM li_y_Page_043.QC.jpg
114e0eb7a1b7032bda5eb133d32146aa
22838020c959998b6b1195ff85b51eaac5ccfaec
9204 F20101220_AACRXY li_y_Page_009thm.jpg
ddcf63d36aecd5a8a9125299cb094712
2e34fd8336057e0cc64808e4f77ed5d47b3eebe4
47923 F20101220_AACRAR li_y_Page_048.jpg
da24c3e0623e1a34d9b307aa87e1ebec
bd9b4119f63ba7cc4d1069fa208200e489421d09
F20101220_AACQVK li_y_Page_019.jpg
36f4398c94a0ba8b5f90a5f75391ff0c
8e9202798e6c866d4e8a78b664ca32c80793dd8b
29668 F20101220_AACQUW li_y_Page_053.QC.jpg
94d949454ca3419391197cb8464db18c
713fb6457957e3a58ab210a307799e8ef5e7c1a4
83552 F20101220_AACRBF li_y_Page_098.jpg
953a422c0b8be4c510869cbf721ba8c3
97d98a29378bc5f96e91d4e0c5441d79b95cccc1
7255 F20101220_AACRZC li_y_Page_061thm.jpg
77e25a417a98b793c5a3bf94b15fdb7b
0ab55f7262beda6cb053b1ec985127efd624ed6b
4228 F20101220_AACRYN li_y_Page_044thm.jpg
18c4e43bb64eaa10372a70989ad16e74
e51de6a98a4468c19f9bfb71e5c067bbb299f2b6
9733 F20101220_AACRXZ li_y_Page_011thm.jpg
07342da738fc92ea542024960471274b
4a4d92c0403a55dd9b15e8f48027fe2f3548e4f3
1521 F20101220_AACQWA li_y_Page_025.txt
aa75106e0abe0c14f2c907bf19724b5e
cd9037b3fd9fdfe00056701b0857ed73bc039d49
F20101220_AACRAS li_y_Page_009.tif
6b5e9224c1813b0f0f061421bf20e78a
8a0d268b28ab88c20b75a8b6046984b110367bd9
2684 F20101220_AACQVL li_y_Page_111thm.jpg
b7ecf98bc02578ec6e6a31a807237808
ddb492a05d2d906c42c4d093441946c73a8dc710
32935 F20101220_AACQUX li_y_Page_084.QC.jpg
39d272226be6d3ca0054327e5d88e0e6
58d8d9637b167405b79548bd8dd129048dbd40db
31959 F20101220_AACRBG li_y_Page_028.jpg
fad494265a9f3e0f48bedff06539e386
5e96d3e63502794b8a70e04bc0933c975a988eea
6536 F20101220_AACRZD li_y_Page_063thm.jpg
bce0f11121938b154487f528bd90a2f6
f2eff901fa224355ddbaf72046bf70ec1008e1ab
18765 F20101220_AACRYO li_y_Page_045.QC.jpg
98aaeacb5a857624c1cbd462fdaed383
7dafa152fcac22fe1be3c72c0ab75f79ce88d97b
1051921 F20101220_AACQWB li_y_Page_021.jp2
225e5f261ebdd7af5e1b72f3acc5ba2a
f0b53d69e96a931c124600c4daebeaf57349125a
907626 F20101220_AACRAT li_y_Page_098.jp2
f4f20037058df5f99884138419e7361f
41cbd9a5d929351868865fe2f47e1b5d3f78a454
49942 F20101220_AACQVM li_y_Page_032.pro
236f809a03a88e594fa2a74f42caa1dc
3558eb2c30851f63164a2f5137a83052c56224db
F20101220_AACQUY li_y_Page_062.jp2
7aa8a0a84f76bbaaeaf7433de008683d
e6ca29a6699225667733745554befa56966f15f7
3640 F20101220_AACRBH li_y_Page_113thm.jpg
c45ed793f2e6dd051798d0895ccbb0fe
a8c75af0d67f211c9d8c0301808c4f2b5e477ff3
4567 F20101220_AACRZE li_y_Page_064thm.jpg
08038e21eae90e58804ea9b90c6a3d54
4de24b07dd6fcf049246dd56a68f0b61a05c4341
6342 F20101220_AACRYP li_y_Page_046thm.jpg
370409caf6661b5e59815a7d1a4829f9
76b2004489e1fbd1dbe43e886e96862affa728e1
1025844 F20101220_AACQWC li_y_Page_034.jp2
cfefea8d1973d1f39381d96faea0142b
abfc6e4334805e336c77eb7eb876737522c388fd
2009 F20101220_AACQVN li_y_Page_041.txt
35e4a52568041f407c13d0be88c52691
ab3a7e903db43866011715dc2e3d03abdefc0420
261199 F20101220_AACQUZ li_y_Page_028.jp2
3ff7a6c15b8b8622c89972a1cc02a2be
e9d82e9277fb92e50ae4b43f44decaf2d908dcb4
621403 F20101220_AACRBI li_y_Page_026.jp2
58f6ca7b1b4155e52c993561b0ea0d62
38cd4d521e3529b08f993e83c5aa64366535e7a1
36217 F20101220_AACRZF li_y_Page_069.QC.jpg
27b9a3054dc547a16e82a62ffc5a2c7c
34dbb3ad1afcdf28abde8ca962052b28e955619a
14272 F20101220_AACRYQ li_y_Page_048.QC.jpg
461a7b13bfa1d028177be7dfae2130a2
46d6ffafa85ffe7be17d2bbf9a2a87cc9e7e03d9
33765 F20101220_AACQWD li_y_Page_065.QC.jpg
f31461fe077ed5726148f24a568e2e8d
5c7e1cfd3da2b9fca82a79a482b6c324ed089349
16071 F20101220_AACRAU li_y_Page_077.QC.jpg
eca62386585eb1b22f68629510635250
949d05b91d55d5482d81ba6e06d5fe5ea776e424
24200 F20101220_AACQVO li_y_Page_134.QC.jpg
c4d35d8550a996d167e26ffae1603cee
a1db4bee054f8c94dbdbdfba42b1b64503998a3f
39463 F20101220_AACRBJ li_y_Page_072.pro
95541c50071251adfea4ca5d7d8d1af7
1a0be7d2eba1c1afde7b569d5499525868dba576
3136 F20101220_AACRZG li_y_Page_071thm.jpg
c6ccaed50a2e7519dc8d67106f9fd36a
2d098bbde8ff9a0155ae9eb4da0254de1eed2683
F20101220_AACRYR li_y_Page_049thm.jpg
eb0bfa7d134f27f5963d81dc6d23fb64
713224011e0d8756490c06381db458f78d4a1bce
97588 F20101220_AACRAV li_y_Page_084.jpg
69602e4fe480113d1c31212ece0c9265
f3ace4bb54475b987b8f107e5148af683ac3a30d
1197 F20101220_AACQVP li_y_Page_087.txt
d06098c05e38f57593ccb1b82dcf73fc
d292e600d63eb46bb7b2d8a93d12497bff85381c
F20101220_AACRBK li_y_Page_098.tif
38cc83e3d04470ff692320001ca21292
5b43defb25570cb30514f32f047b041103ac94bc
43929 F20101220_AACQWE li_y_Page_086.pro
5ce4daac721a4aa26ce53c334ba756fd
4652e71ed7e6103a97b62f4a662f5e167d8dc979
3039 F20101220_AACRZH li_y_Page_072thm.jpg
4f08f66f741641f2e357aac67e564943
6c2e6c389f84f64ee03a2c6395fba5bb6c4146f8
33178 F20101220_AACRYS li_y_Page_050.QC.jpg
7342e44dc5c342ceca021f64ef68f00a
0e49ae6a2caaf5233c861185c5e49e7579fd6755
F20101220_AACRAW li_y_Page_073.tif
520a65e5eb7c3280c2f640477fdd3428
65b60ef43a27c10bb6b9b1dc90962d294e59b0cc
29971 F20101220_AACQVQ li_y_Page_087.pro
a6ddee7792bfc427f1c2d8b2923a82c0
ab517bb0e992d010ae250196f07dfa1e2fe94b0c
14228 F20101220_AACRBL li_y_Page_072.QC.jpg
5e059d02174a80de3a9af2ba3bccc610
dc310113ef008fc3fb975c43a38fdad2541b17d1
8998 F20101220_AACQWF li_y_Page_155.QC.jpg
a95b039c65e5281ac71710b65f47f487
e4ebe6db2fd3284f826506fad0f17e0c06aa523e
3075 F20101220_AACRZI li_y_Page_074thm.jpg
4df7e97508b72f22f39930122d4521cd
18623321d91559c7057c9926f1aa0735b6f40b0e
27733 F20101220_AACRYT li_y_Page_052.QC.jpg
831cd2fa84495c94ac3fca967730e5ba
c53cb593a767e404da68f3650b25187e86b47776
F20101220_AACRAX li_y_Page_091.tif
cb388fdedc7604e57a29e5378c91dc80
2ddd435f4e87a4f1cf4713dba86dc11e581b2083
44329 F20101220_AACQVR li_y_Page_058.jpg
b9d42d73bab1a93b8be98ba1a0623d19
4fb360a3181ec982cb03db4cc153167a6fa43f59
F20101220_AACRCA li_y_Page_086.txt
d0b724f348a244e7411a7345098b5dd4
23f580ea82c1bcbf9164501a03b15d887ca5ac24
1166 F20101220_AACRBM li_y_Page_026.txt
8fb2a35603b62f43b7699c3ce936510a
d1ad2e82a808f1391d141cd0e994811c9a18978e
9598 F20101220_AACQWG li_y_Page_027.QC.jpg
1431549aa58419516a12ee29337ae102
72c37ea512b6fa316c9da73a9854a19addd10390
14109 F20101220_AACRZJ li_y_Page_075.QC.jpg
2e1244d9d4499201eb2d856ec6ae3847
aadd20a9b418dc0a17378d5612742e2e3cb23ad9
7679 F20101220_AACRYU li_y_Page_052thm.jpg
fdd0c0d213a16c39e3559701747547b0
09f6d050526a9bac00bbed1e4928acd7a37b0772
9896 F20101220_AACRAY li_y_Page_049.pro
97f1b5ce296b5ff37e890bbca69658f4
fecd912bb712f9cc5879b8175c4385c72e476b70
65027 F20101220_AACQVS li_y_Page_148.pro
63757cfd42815a4a67e8690e76020a21
cb3b0efe84a3c4feb15e13cddc613fd6a1dfddec
22242 F20101220_AACRCB li_y_Page_064.pro
9c0783a1579e24f2c58875de6396e8b3
20c3ab7ae76ad47c88cdd2962a3a55826cab4af2
8001 F20101220_AACRBN li_y_Page_050thm.jpg
d075491f8187eccfca75077800f7b945
d9a31a0aa4eec466da07701d4f7ccc41d6686316
443271 F20101220_AACQWH li_y_Page_137.jp2
262d51cc2b3ed49cefaac46c1d3216a5
85fda589ed931a66c163cdb693d80b80ba1551ae
3209 F20101220_AACRZK li_y_Page_076thm.jpg
97995cd1f9fc0deb180569f0806fd37a
ae91a85356683b993bdcaf2339c16dc97ebd2589
7993 F20101220_AACRYV li_y_Page_053thm.jpg
3b8783c2538497262eb3e689911bba50
aa61d70bf0cc4c08e1a8e30eaab81a7199a5197b
7464 F20101220_AACRAZ li_y_Page_086thm.jpg
a7291afba327fc20652087e960bd61ac
a171b3cc98b9b38905ea15e3ad07e44675089984
39274 F20101220_AACQVT li_y_Page_074.pro
2a8d9ed6d12905828e06872f81d9fda8
972c9a49541cfd4f13e18d351403c5840f7d99b0
F20101220_AACRCC li_y_Page_016.jp2
535a2fd42703e4a09b3506caf1c5acac
e917ced12f33ac8b33ec599d6506209bc80b3791
13949 F20101220_AACRBO li_y_Page_073.QC.jpg
0eed4576d2e211eb4cd49e878ce52e46
cafc86b38829b78a6b17d3c73f122b1b228ca25c
138793 F20101220_AACQWI li_y_Page_142.jpg
9a36c9d0cb67324886f7e66c9780f077
976f189f536d4531e8593daa57e33903f40037bc
5431 F20101220_AACRZL li_y_Page_078thm.jpg
b9743feabd58c23603d95c137ecb1007
b46d7f090e0204cbc7165adc9027b585413c4414
32023 F20101220_AACRYW li_y_Page_054.QC.jpg
8c4e7cc4c227391f547842c6c7618e1b
70891daa9ca39edb9e26dba4a0fcc02c677a401f
17475 F20101220_AACQVU li_y_Page_040.QC.jpg
e8a4ec74c0d9acae5ebe580eb60faa7a
e6f4002de61d968b4ead5eaf81ed5517c3a4943b
9174 F20101220_AACRCD li_y_Page_152thm.jpg
31d5309d15ebb86b25148b9745c2da83
b6d89db8c492d84ca7ce8c82ff88e3f17ef8d9c2
36570 F20101220_AACRBP li_y_Page_021.QC.jpg
17d5b6796a2e3fcb1bf89a18779af19c
4791fcb2ac93b8b63fe59241d518559cf5f885fb
37828 F20101220_AACQWJ li_y_Page_020.QC.jpg
8b8a3225bd800f64bc312c2d4889c9d4
1b55efc9de2a9312ff38a1e10cefd562d4765044
8011 F20101220_AACRZM li_y_Page_084thm.jpg
f7d1ffa17bf4b119e868f8230f1dbdea
73fc5b64e15b6926d8a9db58ccfed97d1264db0c
4682 F20101220_AACRYX li_y_Page_056thm.jpg
db39308fcc9ec88760f51b5023f57350
977b35b875bf73631eaa7acc8d6fe2f97684b25e
F20101220_AACQVV li_y_Page_044.tif
d7c89bd8b5fd1ada2172693d63d42498
44b7bc588e572bb353a54c7672bd333bcbad2728
69619 F20101220_AACRCE li_y_Page_041.jpg
5424e635f17d069d62634099f821de85
0f4d96e72d7e34d4ae46a9ec8926dbc5a6fe1988
8185 F20101220_AACRBQ li_y_Page_093thm.jpg
54640153b796cfef418d425fd626f5bf
87e244f585b383e150b751c81fd4b5c21ef8dfbc
477 F20101220_AACQWK li_y_Page_155.txt
d2a5ee170063426a1f5938956f5aa643
9c99c4aafb560ada1e57bc87efee9bf4ca5aab68
3761 F20101220_AACRYY li_y_Page_058thm.jpg
e7518559c6c680f39cd95b1e3a9c3c95
27931ff4ae61dabba2172cbe5789bc4dfed16e1b
20012 F20101220_AACQVW li_y_Page_042.QC.jpg
62b5054db195bdb3984ff48c16c73a82
0b0160ad17cccfbe3fd35a99aa9d78476bf5cf2b
973725 F20101220_AACRCF li_y_Page_102.jp2
19cbed226bf94b02d6e7ed71351cbf4c
69e389e4b2fa66a0590739be3405df6caf59a2f9
9659 F20101220_AACRBR li_y_Page_132.pro
4dcd3ef2c5698e99768a8975a83fa95f
38c780a6e308a446eb8d9d174853eec7067a9521
5153 F20101220_AACRZN li_y_Page_087thm.jpg
65f8e36e5761cfa971096b2f2245e5e7
9697c13bce3f036e1ad2a0ac8b85facac219ef03
19644 F20101220_AACRYZ li_y_Page_059.QC.jpg
4d0ad030620b9bf2bb737e9aaad77937
038ef2e4f518df5db46e911194cd37efa44240c9
F20101220_AACQVX li_y_Page_012.tif
d9b9d09396992d14e6c24f8e7dbd06e6
bd03e19afdb6de05caddc4599d541b16b6eb2412
F20101220_AACRCG li_y_Page_037.tif
0aa8a781ac9bb6e769450058bc7defaa
784d19b5ee7f5b25acd078b2f68ce2865c93731a
24897 F20101220_AACQXA li_y_Page_119.QC.jpg
a4c4681b0284475b15779426740c53b8
3886553fc2b25a56b40997b3b9e9fa9c095b7796
41766 F20101220_AACRBS li_y_Page_011.QC.jpg
92e316f102757854ffee181103678f83
144806f726c4a80d78ca3675c5ec185bbbce8c5e
F20101220_AACQWL li_y_Page_066.tif
c2c9f0b3bea95284d5f41064e0f9ca8c
e34e2a718baf53d2ad9582f933d88045406c7af8
31300 F20101220_AACRZO li_y_Page_094.QC.jpg
9d7d6e63303e1cd9486d8aba8684c82e
4de145cd4199595943f76eab171a05908c6eef21
1256 F20101220_AACQVY li_y_Page_099.txt
ccd4f3646601ff8037d56e2910664d6a
9aa4f02a7b8eb33803e2b549e9291db168e28d11
F20101220_AACRCH li_y_Page_129.tif
8e41ca755841d4dbe6b7f22d94127838
4fbab428b237e721405834369af01b4081c1d2a0
20768 F20101220_AACQXB li_y_Page_128.QC.jpg
5c9b090a14625f3a3f182fa8b01f638d
885889eee7e379fc7a9b048e05339f678556b5d8
976669 F20101220_AACRBT li_y_Page_085.jp2
3a5b2acc9173c8228f07fce9a13e56be
29859878736bcd4ade4a85188ceba81642f83a09
8473 F20101220_AACQWM li_y_Page_055thm.jpg
de82e0a7a61ceebd96ebc4a0689e8552
6613742ad5ef4548d5d4176ee597820ce86cf195
7241 F20101220_AACRZP li_y_Page_098thm.jpg
1e56d827e3d8c76562b42e7738733615
7c03def436f60903a10fc2b0aa756dd0a24174b5
50175 F20101220_AACQVZ li_y_Page_108.pro
f01462d3b13f94e4823b7b1e41bf6c2c
13d7315488c0c2cef9d57ecdc517c77d1cb5ef86
124481 F20101220_AACRCI li_y_Page_010.jpg
48d48506387b8f514a5e774e71233a10
510e163d687bb6cc3e6462432361a492eb4c9ebe
68256 F20101220_AACQXC li_y_Page_152.pro
46c1ad3fcbc98ce2e710b7afcd7780c8
c9221f7701c7f46b0c6f1a666ea5179a9f78f440
101125 F20101220_AACRBU li_y_Page_016.jpg
96f03892faf5063dd5a21dc9196cb873
da39fbc4e99dde53912433efcb3745d3c4365581
27294 F20101220_AACQWN li_y_Page_060.QC.jpg
c9dc1b99b5a89458cf103e448133d45e
256fdf27fd674fe0edb3fa03ea995cf0f6f05ba5
5856 F20101220_AACRZQ li_y_Page_099thm.jpg
913fd09e9f179e7aca04d589c4cbf9a4
0bf1384222eccea2d1c21497d6d904aad58965de
4034 F20101220_AACRCJ li_y_Page_116thm.jpg
044b3f1068430da9c8557c2f8893e505
582f4a1b01ad5f87264825e5f8bbdfbd5e6c035d
F20101220_AACQXD li_y_Page_021.tif
a00dff00e8453183e10e59f3b82fc447
03b3595cb04bf4b34529434e7f14f8914c88eee1
68176 F20101220_AACQWO li_y_Page_135.jpg
fa5168b00ba5691759ded4bcd6bcccfd
87d9d344910baae0c8333ba6159846f3709c04bb
F20101220_AACRZR li_y_Page_103.QC.jpg
bc9aa1b1640b3a8d691a4ffa314828d0
232cd9dd3760a62fcf45b5e27ff314a2f7a2c8a2
108804 F20101220_AACRCK li_y_Page_021.jpg
79161e3c358fc9421525f71f5d697e1b
f69b6a0fbeee71374635409d17b8838e6252064d
F20101220_AACQXE li_y_Page_155thm.jpg
707f55a10b33f6e8793219bdd564594d
cc487fd4754328d520a193d697c1cb1b8bc51c31
F20101220_AACRBV li_y_Page_126.tif
202fae101c83d4fb76ffaf9a358ef374
3b13a6de323f72f7b143719e2a6c84ccaa607835
F20101220_AACQWP li_y_Page_060.jp2
dd3d1f21dfa7b00a041eb6a8b4be783e
bb65232aff1796eeaa900801eec08d4fee8f72cf
8208 F20101220_AACRZS li_y_Page_104thm.jpg
e1625c8f4fe0eb97a78b9c16d1b02fb6
f370cdf1a6f07cc598076b36734c78b66d84cdd1
2100 F20101220_AACRCL li_y_Page_015.txt
ce607033987981b8a07482dc4ad5c7dd
6bb4bbf62476fcf69c065b86104f1df451648f8a
2784 F20101220_AACQXF li_y_Page_145.txt
8eb58cddc39681b72e8b9d0c47fb795e
07e050f4fd63fa66bfa4aef4f421ab2a0b9e3891
1051940 F20101220_AACRBW li_y_Page_149.jp2
e619a7c0f28185b90b592c585c4f9665
4e2bf070bf7590ab4f76c11198ea9b2fc21b7e22
357 F20101220_AACQWQ li_y_Page_154.txt
fa11dd00d673d740c493deb52f278946
0bd4e1545f79c9ffd17687e48b956e77a11f89c6
30607 F20101220_AACRZT li_y_Page_106.QC.jpg
21ff3bc0c9afb6051c034207f2886486
a3c870a6574cd2eb63a7b9fab7ed9ed0b2ade9f0
F20101220_AACRDA li_y_Page_089.tif
bf73451a4488a56b6ffa17532f1a2e12
d5fe1c96c7dafac6ec386a23458870e8bbf5357d
904588 F20101220_AACRCM li_y_Page_075.jp2
fe93b278aca94449cc00bc7c0be05557
da4f8a957761e10dbb847320fd13144c48203316
300570 F20101220_AACQXG li_y_Page_029.jp2
4a812c0c169fdbb5be161ac9079d1482
b7b26807b7c4eae6408b3bc0a58fe8f2cdd7ecd3
744 F20101220_AACRBX li_y_Page_058.txt
64bad6c70b430a1ac6f44ac2c9742f0c
16777e8cd22d52f33a5e497d48f98e45a5006c69
1051899 F20101220_AACQWR li_y_Page_105.jp2
7a17bc651821237c8446f135967adc57
0e32b79616410a4133cee85008b2d50771785b7a
33863 F20101220_AACRZU li_y_Page_107.QC.jpg
5cc46258e3f6f9ce0dfb87867e91609e
129c6f204b1dbf5e2df8a6b60bdeb6de184cc7be
107527 F20101220_AACRCN li_y_Page_062.jpg
2e2bdd44aecc927d7277721cc3efb616
979bafdc88e6107ce1f3f2a59348622bc90118f9
1926 F20101220_AACQXH li_y_Page_039.txt
c3e75a6b618fbd8c6f823ec8a8312e75
1e0ef09296e2e0468221ac6dbd1cb633ddf7335e
F20101220_AACRBY li_y_Page_088.txt
a698b80000f55e2305dba1b26d70e76f
6e623072cc036ed0b00bbe22002a572003dc7669
42099 F20101220_AACQWS li_y_Page_053.pro
caa4ea4753278fa16a4772592d0cdae9
218821c7541edd8e6a87c4017ddca62442dc54a5
34032 F20101220_AACRDB li_y_Page_138.QC.jpg
c9e0e4a9b9a0556eb8f901f625814e32
d964882f1f594dac1f8aed5e726ca5f4f499fadf
8275 F20101220_AACRZV li_y_Page_107thm.jpg
3ae42607773eb74409987af3e54ade56
b57526386a2d43c92ca247fc0970593df37e2d18
1647 F20101220_AACRCO li_y_Page_098.txt
83dd357dee07f4c95099343554849120
e9d4874c5ba47246d1333cf4b44f27853590d129
F20101220_AACQXI li_y_Page_092.tif
932ab04fcfababd088dea6db575a32da
a19b73d02f54d7a480385de600d90a0a1ca77667
688 F20101220_AACRBZ li_y_Page_029.txt
585d1f946c84c0213592002c8bcc6184
8423bf06e18c9c562e16a361b6490c15c7566c5d
F20101220_AACQWT li_y_Page_022.jp2
355b2baf28e7476c5a5fbcd603d7b949
77ad16c38707a96bf00c2e3fba5b69124be74c5d
893691 F20101220_AACRDC li_y_Page_037.jp2
81c6739d4d8c86b1cec3c95f79a0b81b
101eada95f10685fabad423ba1a87f6417357d86
33971 F20101220_AACRZW li_y_Page_108.QC.jpg
18d39984253ad6645e8ab431f0847655
16a37399f6c04587ce4e9e315978929fc65137a4
1051953 F20101220_AACRCP li_y_Page_024.jp2
2420c8faee94f8d141c4a1fa7d8f20f2
5d8c77144c583503f4f983c0eb9449df177cb2bc
846062 F20101220_AACQXJ li_y_Page_025.jp2
51c52983f098a4f89107d3ffe2ffb740
ac6b776a6e40e08f64654ac1f09b42b65ed26284
841540 F20101220_AACQWU li_y_Page_126.jp2
d7bc33ec24f9b65c279d69e370dc2cd7
943114413ffe70bb9a0722a445a84e1484265bf5
F20101220_AACRDD li_y_Page_104.jp2
d886459b2113389249b4314e857a50df
3241266e1c61c18be5e51548dafaa1cb86663607
8554 F20101220_AACRZX li_y_Page_108thm.jpg
db617a6c2e8feb0ed02d1f3a55aa2036
3d23b9e325edd5a6bc7cffd0bd9bb6fb2a519442
624825 F20101220_AACRCQ li_y_Page_044.jp2
5e78dae215688bb7acd754610f0c9644
c05bad172db75a5e4e8d3d4d47b5ad55a8aac0be
491 F20101220_AACQXK li_y_Page_140.txt
c6ca50139da7c71b400f44e3d2efc5e8
7fd63c97c377508dfe6d20b92957741b16a287eb
681865 F20101220_AACQWV li_y_Page_087.jp2
9ec77d6f227ae40662011d13224e4bf7
8bf3e8effd4d517917cc5606e3efcf3bd8e9b8c7
F20101220_AACRDE li_y_Page_045.tif
4e57cd68af25a1abbf86a038578e8c8d
eb1a583fe9bd07dc68021fc4d3f1ef0219fabb11







PHARMACOKINETIC S/PHARMACODYNAMIC S OF VORICONAZOLE


By

YANJUN LI
















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008

































O 2008 Yanjun Li




































To my beloved husband, daughters, and parents, for their love









ACKNOWLEDGMENTS

I express my deep appreciation and grateful thanks to Dr. Hartmut Derendorf for his

intelligent guidance and generous support throughout my Ph.D. study. He gave me the

opportunity of being part of his group exploring the opportunities in research area of

pharmacokinetics and pharmacodynamics. The knowledge I learned from him benefits me and

will be cherished in my entire lifetime.

Thanks go to Dr. Cornelius J. Clancy and Dr. Minh-Hong Nguyen for their broad

knowledge in microbiology and infectious diseases, intelligent guidance and generous support in

the lab space and equipment necessary for my experiments. It has really been a wonderful

experience to learn knowledge in their laboratories. Without their help the pharmacodynamic

studies would not have been performed. I thank Dr. Veronika Butterweck for her support and

guidance during the animal studies and in my projects. I offer my grateful thanks to the members

of my supervisory committee, Dr. Guenther Hochhaus and Dr. Kenneth Rand for their support

and valuable advice.

I thank Stephan, Jian, Vipul, Sab for the study discussion and help. I thank the

administrative people of the Department of Pharmaceutics, Mr. Marty Rhoden, Mrs. Patricia

Khan, and Mrs. Robin Keirnan-Sanchez, for their technical support. I also would like to extend

my thanks to all faculties of the Department of Pharmaceutics, staffs, graduate students, and

post-doc fellows who were my colleagues for their support and friendship.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. ...............4.....


LI ST OF T ABLE S ............ ..... ._ ...............8....


LIST OF FIGURES .............. ...............10....


AB S TRAC T ............._. .......... ..............._ 13...


CHAPTER


1 INTRODUCTION ................. ...............15.......... ......


Background ................. ...............15.................
Voriconazole................. .............1
Mechanisms of Action ................. ...............16........... ....

Drug Stability .............. ........ ...............17
Animal Pharmacokinetic Studies .............. ...............17....
Human Pharmacokinetic Shtdies................... ..............1
In Vitro and In Vivo Pharmacodynamic Studies............... ...............18
Clinical Efficacy ................. ...............20..._._._ ......
Microdialysis .................... ...............21.
Principle of Microdialysis .............. ...............22....
Features of Microdialysis ................. ...............22................
Application in PK/PD Studies ................. ...............23................
The PK/PD Approaches............... ...............2
Hypothesis and Obj ectives ................. ...............25.......... ....

2 MEASUREMENT OF VORICONAZOLE ACTIVITY AGAINST CANDIDA
ISOLATES USINTG TIME-KILL METHODS VALIDATED BY HIGH
PERFORMANCE LIQUID CHROMATOGRAPHY ........................... ........._.._. ...30


Back ground ................. ...............3_ 0...._._._....
Specific Aims............... ...............3 0.
Materials and Methods .............. ...............30....
Antifungal Agents .............. ...............30....
Test Isolates ........._..._... ...... ._. ._. ...............30.....
Antifungal Susceptibility Testing ........._..._._........_ ...............31....
Anti fungal Carry over. ........._..._.._ ...............3_ 1...._._._....
Time-kill Experiments ........._..._... ...._._._ ...............32.....
Postantifungal Effect (PAFE) Experiments....................... ... ..... ...........3
HPLC for Determination of Voriconazole in Stock Solution and RPMI Medium .........32
Calibration Curves for Voriconazole in Stock Solution and RPMI Medium ................3 3
Statistical Analysis .............. ...............33....
Re sults............ ..... ._ ...............34....












Stability of Voriconazole in Stock Solution and RPMI Medium .............. ..............34
Minimum Inhibitory Concentration (MIC) .............. ...............34....
Time-kills and Postantifungal Effect (PAFE) .............. ...............34....
Discussion ................. ...............36.................


3 APPLYING PHARMAC OKINETIC/PHARMAC DYNAMIC MATHEMATICAL
MODEL ACCURATELY DESCRIBES THE ACTIVITY OF VORICONAZOLE
AGAINST CANDIDA SPP. IN VITRO .............. ...............50....


Background ................. ...............50.................
Specific Aim .............. ...............51....
M materials and M ethods .................. ........ ...... .........5
Mathematical Modelling of Time-kill Data ................. ............ ......... ........ .......5
Simulations of Expected Time-kill Curves Using Human PK Data..........._._... .............5 1
Statistical Analysis .............. ...............52....
R esults........ ........ .....__ ...........__ .. .............._ .... ...................5
An Adapted Sigmoidal Emax Model Provides the Best Fit for Voriconazole Time-
kill Data against Candida Isolates.............. ..... ...... ....... ...............5
Human PK Data for Voriconazole Can Be Used to Simulate Expected Time-kill ........54
Discussion............... ...............5


4 PHARMACODYNAMIC STUDY IN DYNAMIC SYSTEM FOR DESCRIBING THE
ACTIVITY OF VORICONAZOLE AGAINST CANDIDA SPP. IN VITRO ................... .....62


Back ground ................. ...............62.......... ......
Specific Aim .............. ...............62....
M materials and M ethods .............. ...............63....
Anti fungal Agents .............. ...............63....
Test Isolates ................. ...............63.......... ......
Softw ares .............. ...............63...

Dynamic Model Design............... .......................6
Time-kill Experiments in the Dynamic Model ................. ...............65...............
Re sults ................ ...............66.................
Discussion ................. ...............69.................


5 IN VITRO MICRODIALYSIS OF VORICONAZOLE .........._._ ......_.... .................8 1


Back ground ................. ...............8.. 1..............
Specific Aims............... ...............8 1.
Materials and Methods .............. ...............82....
M materials ................ ...............82.................
M ethod s ................. ...............8.. 2..............
Recovery ............... ... .. .... ........ ...............82.......
Extraction efficiency method (EE) ................. ...............82................
Retrodialysis method (RD)............... ...............83..
Microdialysis experiments .............. ...............84...
HPLC for determination of voriconazole ......___ ..... ... .__. .. ...._.__.........8












Calibration curves for voriconazole .............. ...............85....
Re sults.........__ _........ ._ .. ...._._ ... ... ...... ........8
Stability of Voriconazole in Lactate Ringer' s Solution and Plasma .........._... ..............86
Protein Binding............... ...............86
Discussion ................. ...............86.................


6 INT VIVO PHARMACOKINETIC STUDIES OF VORICONAZOLE INT RATS .................91


Back ground ................. ...............9.. 1..............
Specific Aims............... ...............9 1.
M materials and M ethods .............. ...............92....

Reagents and Equipment ................. ...............92................
Animal s................ ...............92

Experimental Design .............. ...............92....
Anesthetic Procedure ................. ...............93.................
Antiseptic Procedure .............. ...............93....
Blood Samples............... ...............93
M icrodialysis .............. ...............94....
Probe calibration .............. ...............94....
M uscle microdialysis............... .............9
Data Analysis............... ....... ... .......9
Noncompartmental pharmacokinetic analysis .............. ...............96....
Compartmental pharmacokinetic analysis and modeling............... ................9
Re sults........._..... ...._... ...............102....
Probe Recovery ................ .... ... .... ... ..... .... ........0
Individual Pharmacokinetic Analysis of Total Voriconazole in Plasma after
Intravenous Bolus Administration ......__....._.__._ ......._._. ............10
Noncompartmental pharmacokinetic analysis .............. ...............103....
Compartmental pharmacokinetic analysis .............. ...............103..
One-compartment model with linear elimination analysis............... .................0
Two-compartment model with linear elimination analysis............... ...............10
One-compartment model with nonlinear elimination .............. ....................10
Two-compartment model with nonlinear elimination.................................. 10
Individual Pharmacokinetic Analysis of Unbound Voriconazole in Muscle after
Intravenous Bolus Administration ......__....._.__._ ......._._. ............10
Noncompartmental pharmacokinetic analysis .............. ...............107....
Compartmental pharmacokinetic analysis .............. ...............107....
Two-compartment model with nonlinear elimination.................. ............10
Pharmacokinetic Analysis of Average Total Voriconazole Data in Plasma and
Unbound Voriconazole Data in Muscle after Intravenous Bolus Administration .....108
Discussion ........._.___..... ._ __ ...............109....


7 CONCLUSIONS .............. ...............138....


LIST OF REFERENCES ........._.___..... .___ ...............141....


BIOGRAPHICAL SKETCH ........._.___..... .__. ...............155....










LIST OF TABLES


Table page

1-1 Pharmacokinetic parameters of voriconazole in mouse, rat, rabbit, guinea pig and
dog following single and multiple administration by oral and intravenous routes........... .26

1-2 Pharmacokinetic (PK) properties of triazole antifungals ................. ................. ......27

2-1 Percentage (%) of voriconazole maintenance in stock solution and RPMI medium at
different temperatures .............. ...............3 8....

2-2 Minimum inhibitory concentration (MIC) of voriconazole against Candida isolates .......38

2-3 Can2dida albicans ATCC90029 in constant concentration experiments ................... .........39

2-4 Can2dida albicans SC5314 in constant concentration experiments ................. ................39

2-5 Can2dida glabrata 1 in constant concentration experiments .............. ....................4

2-6 Can2dida glabrata 2 in constant concentration experiments .............. ....................4

2-7 Candida parappilsiloi 1 in constant concentration experiments .............. ...................41

2-8 Candida parappilsiloi 2 in constant concentration experiments .............. ...................41

2-9 Summarized voriconazole time-kill data against Candida isolates............... ................4

2-10 Candida albicans ATCC90029 in PAFE experiments ......____ ........_ ..............42

2-11 Can2dida glabrata 2 in PAFE experiments .....__.....___ ..........._ ..........4

2-12 Can2dida glabrata 1 in PAFE experiments............... ..............4

2-13 Candida parappilsiloi 1 in PAFE experiments............... ..............4

2-14 Can2dida parapsilosis 2 in PAFE experiments .....__.....___ ........... ............4

3-1 Pharmacodynamic parameters and goodness of fit criteria against Candida isolates........57

3-2 Steady-state pharmacokinetic parameters in plasma as calculated by two-
compartment model analysis............... ...............58

4-1 Can2dida albicans ATCC90029 in changing concentration experiments ..........................71

4-2 Candida glabrata 1 in changing concentration experiments ................. ............. .......72

4-3 Candida glabrata 2 in changing concentration experiments ................. ............. .......73










4-4 Candida parapsilosis 1 in changing concentration experiments .............. ...................74

4-5 Candida parapsilosis 2 in changing concentration experiments .............. ...................75

4-6 Pharmacodynamic parameters and goodness of fit criteria against Candida isolates in
the dynamic infection model ................. ...............76........._....

5-1 List of materials ................. ........._.... ...._.. ......... ...._... ....._. ...88

5-2 Extraction efficiency (EE) method for measuring voriconazole recovery ........................89

5-3 Retrodialysis (RD) method for measuring voriconazole recovery .............. ..................89

5-4 Rat plasma protein binding data of voriconazole measured by in vitro microdialysis ......90

5-5 Human plasma protein binding data of voriconazole measured by in vitro
m icrodialysis............... .............9

6-1 Recovery data of voriconazole using retrodialysis (RD) method in rats muscle............. 112

6-2 Total plasma voriconazole (5 mg/kg) individual noncompartmental PK analysis..........1 13

6-3 Total plasma voriconazole (10 mg/kg) individual noncompartmental PK analysis........114

6-4 Comparison of obj ective function value (OFV) and Akaike information criterion
(AIC) value from rats total plasma voriconazole PK analysis using different models....115

6-5 One-compartment model with linear elimination for rat total plasma voriconazole PK
analysis............... ...............11

6-6 Two-compartment model with linear elimination for rat total plasma voriconazole
PK analy si s ................ ...............116....._ .__....

6-7 One-compartment with non-linear elimination model for rat total plasma
vori conazol e PK analy si s ................. ................. 1...._ ._ 17....

6-8 Two-compartment with non-linear elimination model for rat total plasma
vori conazol e PK analy si s ................. ................. 1...._ ._ 17....

6-9 Unbound muscle voriconazole (5 mg/kg) individual noncompartmental
pharmacokinetics analysis ................. ...............118......... ......

6-10 Unbound muscle voriconazole (10 mg/kg) individual noncompartmental
pharmacokinetics analysis ................. ...............119......... ......

6-11 Rat total plasma and unbound muscle voriconazole PK analysis using two-
compartment with non-linear elimination model ................. ..............................120










LIST OF FIGURES


FiMr page

1-1 Structural relationship among azole drugs................. ...............28

1-2 Plasma voriconazole concentration-time profiles following i. v. dosing and following
oral dosingof voriconazole............... ..........2

2-1 Susceptibility of voriconazole against Candida isolates ........._.__........_ .........._....45

2-2 Culture flasks for voriconazole against Candida isolates in vitro............... .................4

2-3 Time-kill curves for voriconazole against Candida isolates ........._._...... .._._...........46

2-4 Voriconazole did not demonstrate PAFEs ................. ...............47...............

2-5 Chromatogram of voriconazole which concentration was determined from peak area.....48

2-6 Voriconazole concentrations in culture media throughout the duration of time-kill
experiments by HPLC ...........__......___ ...............48....

2-7 Voriconazole concentrations in PAFE experiments by HPLC............ ..__ .........__ ....49

3-1 Plasma VOR concentration-time profiles simulated with a two-compartment PK
m odel ................ ...............59........ ......

3-2 Fitted time-kill curves derived using the mathematical model for constant
concentrations of voriconazole ................. ...............60._._. ......

3-3 Simulations of candidal time-kills and plasma voriconazole concentration-time
profiles. ............. ...............61.....

4-1 Concentration elimination curve of voriconazole in the dynamic infection model...........77

4-2 Time-kill curves from the dynamic in vitro model ....._.__._ .... ... .___ ........_........78

4-3 Fitted time-kill curves were derived by our mathematical model for changing
concentrations of voriconazole. ............. ...............79.....

4-4 Using parameters from dynamic models to simulate candidal time-kills and plasma
voriconazole concentration-time profiles .............. ...............80....

6-1 Cascade pharmacokinetic one-compartment model with non-linear elimination of
voriconazole scheme ........._.__........__. ...............121...

6-2 Cascade pharmacokinetic two-compartment model with non-linear elimination of
voriconazole scheme ........._.__........__. ...............121...










6-3 Dosage of 5 mg/kg i. v. bolus: total voriconazole concentration in rat plasma. .............122

6-4 Dosage of 10 mg/kg i. v. bolus: total voriconazole concentration in rat plasma ..............123

6-5 The PK plots using one-compartment with linear elimination model for PK analysis
of rat total plasma voriconazole data ................ ...............124..............

6-6 Goodness of fit plots for observed vs predicted concentrations using one-
compartment with linear elimination model for PK analysis of rat total plasma
voriconazole concentration data. ............. ...............125....

6-7 Goodness of fit plots for residuals using one-compartment with linear elimination
model for PK analysis of rat total plasma voriconazole concentration data. ........._._......125

6-8 The PK plots using two-compartment with linear elimination model for PK analysis
of rat total plasma voriconazole data. ................ ........................ ..............126

6-9 Goodness of fit plots for observed vs predicted concentrations using two -
compartment with linear elimination model for PK analysis of rat total plasma
voriconazole concentration data. ............. ...............127....

6-10 Goodness of fit plots for residuals using two-compartment with linear elimination
model for PK analysis of rat total plasma voriconazole concentration data. ........._._......127

6-11 The PK plots using one-compartment with non-linear elimination model for PK
analysis of rat total plasma voriconazole data ................. ........__. ....___.........128

6-12 Goodness of fit plots for observed vs predicted concentrations using one-
compartment with non-linear elimination model for PK analysis of rat total plasma
voriconazole concentration data. ............. ...............129....

6-13 Goodness of fit plots for residuals using one-compartment with non-linear
elimination model for PK analysis of rat total plasma voriconazole concentration
data ................. ...............129................

6-14 The PK plots using two-compartment with non-linear elimination model for PK
analysis of rat total plasma voriconazole data ................. ........__. ....___.........130

6-15 Goodness of fit plots for observed vs predicted concentrations using two-
compartment with non-linear elimination model for PK analysis of rat total plasma
voriconazole concentration data. ............. ...............131....

6-16 Goodness of fit plots for residuals using two-compartment with non-linear
elimination model for PK analysis of rat total plasma voriconazole concentration
d ata. ................. ...............13. 1..............

6-17 Dosage of 5 mg/kg i. v. bolus: unbound voriconazole concentration in rat muscle .........132










6-18 Dosage of 10 mg/kg i. v. bolus: unbound voriconazole concentration in rat muscle .......133

6-19 The PK plots using two-compartment with non-linear elimination model for analysis
of rat total plasma and unbound muscle voriconazole data. ................ ........_._ .......134

6-20 Goodness of fit plots for observed vs predicted concentrations using two-
compartment with non-linear elimination model for PK analysis of rat total plasma
and unbound muscle voriconazole concentration data. ............. ....................13

6-21 Goodness of fit plots for residuals using two-compartment with non-linear
elimination model for PK analysis of rat total plasma and unbound muscle
voriconazole concentration data. ............. ...............135....

6-22 Dosage of 5 mg/kg i. v. bolus in rat: average plasma and muscle PK data analysis
using two-compartment with non-linear elimination model ................. .....................136

6-23 Dosage of 10 mg/kg i. v. bolus in rat: average plasma and muscle PK data analysis
using two-compartment with non-linear elimination model ................. .....................137









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PHARMACOKINETIC S/PHARMACODYNAMIC S OF VORICONAZOLE

By

Yanjun Li

August 2008

Chair: Hartmut Derendorf
Major: Pharmaceutical Sciences

Candida spp. is the most common cause of opportunistic mycoses worldwide, especially

the invasive opportunistic mycotic infection. Oropharyngeal candidiasis is the most common

opportunistic infection associated with AIDS. Candidemia is the fourth most common

bloodstream infection in the developed world and is associated with mortality rates of up to 3 8%

despite treatment with available antifungals. Voriconazole, a triazole agent that is effective in

treating candidiasis and aspergillosis, inhibits ergosterol synthesis by blocking the action of 14a-

demethylase.

It has also lowered toxicity compared with previous antifungal agents; however, it is used

to cure candidiasis inside the human body after oral administration or i.v. inj section. Empiric

antifungal therapy is initiated as the first measure dealing with fungal infection, and there are no

suitable mathematical models of voriconazole antifungal activity established for rational

guidance to maximize efficiency and minimize toxicity.

The aim of these studi es was to develop an optimal pharmacokineti c/pharmacodynami c

(PK/PD) model to predict clinical outcome and improve the effectiveness of voriconazole. The

PK/PD model integrates the pharmacokinetics of unbound voriconazole with in vitro antifungal

activity against certain Candida strains for rational dosing of this agent.









In these studies, we measured voriconazole activity against Candida isolates using time-

kill methods validated by high performance liquid chromatography in vitro and a

pharmacokinetic/pharmacodynamic mathematical model based on time-kill curves against

different Candida strains was used to accurately describe the antifungal activity of voriconazole.

Using this model, pharmacodynamic studies in dynamic system for describing the activity of

voriconazole against Candida spp. were conducted. Moreover, in vitro microdialysis was

performed to determine protein binding of voriconazole in both rat and human plasma. Also, in

vivo microdialysis in Wistar rats after intravenous administration of voriconazole was conducted

to evaluate unbound muscle concentrations of voriconazole and to compare the free tissue

concentration to free plasma concentration. We investigated the voriconazole PK profile by

analyzing the total concentrations in plasma and unbound concentrations in muscle and

developed a population PK model to fit the voriconazole PK data in rats using NONMEM.

These studies provided new insights to improve the drug clinical efficacy by using the

developed voriconazole PK/PD model.









CHAPTER 1
INTTRODUCTION

Background

In the past two decades, the incidence of invasive fungal infections, especially in

immunocompromised patients, such as AIDS patients have risen significantly. Invasive fungal

infections are growing in frequency and optimal selection and dosing of antifungal agents are

important. Agents administered for invasive infections are amphotericin B, flucytosine, and azole

antifungals. Azoles, a class of organic compounds having a five-membered heterocyclic ring

with two double bonds, have gained attention as useful drugs with antifungal activity in vitro

against a broad spectrum of fungal pathogens. The azole antifungal agents inhibit the synthesis

of ergosterol by blocking the action of cytochrome P450 14a-demethylase (P45014DM) [1-4]. The

structure of six main azole drugs including ketoconazole, fluconazole, itraconazole,

voriconazole, ravuconazole, and posaconazole is shown in Figure 1-1 15, 6]. Antimicrobial PD

describes the relationship between drug exposure and treatment efficacy. Therapeutic outcome

predictions based upon these PD relationships have correlated well in treatment against both

susceptible and resistant pathogens. The potential value of using PK/PD parameters as guides for

establishing optimal dosing regimens for new and old drugs, for new emerging pathogens and

resistant organisms, for setting susceptibility breakpoints, and for reducing the cost of drug

development should make the continuing search for the therapeutic rationale of antifungal dosing

of animals and humans worthwhile [11 71

Voriconazole

Voriconazole (UK-109,496) is a triazole that is structurally related to fluconazole (Figure

1-1) [5 8. It is developed by Pfizer Pharmaceuticals (http://www.pfizer. com) and its clinical use

was approved by the Food and Drug Administration (FDA) in May 2002. The trade name of









voriconazole is VfendTM [9]. Voriconazole recently has gained increasing attention as a new class

of azole antifungal drug with its lower toxicity and expanded antifungal activity against Can2dida

spp. compared with previous antifungal therapies, and also with its advantage of high activity

against Aspergillus specieS [10-12]. Voriconazole is fungistatic and exhibits no PAFE against

Candida albicanS [13-15]. Time-kill and PAFE data are limited against C. glabrata and do not

exist against C. parapsilosis isolates. Moreover, standard time-kill and PAFE methodologies,

although widely used, have not been validated for voriconazole or other antifungals by direct

measurements of drug concentrations [16]

Mechanisms of Action

All azole antifungal agents work principally by inhibition of P45014DM. This enzyme is

in the sterol biosynthesis pathway that leads from lanosterol to ergosterol 16, 17-19]. COmpared to

fluconazole, voriconazole inhibits P45014DM to a greater extend. This inhibition is dose-

dependent [20-22]. Voriconazole also has an enhanced antifungal spectrum that includes

filamentous fungi [23-26]. Voriconazole also inhibits 24-methylene dihydrolanasterol

demethylation in certain yeast and filamentous fungi. Other antifungal effects of azole

compounds have been proposed and include: inhibition of endogenous respiration, interaction

with membrane phospholipids, and inhibition of the transformation of yeasts to mycelial forms

[19]. Other mechanisms may involve inhibition of purine uptake and impairment of triglyceride

and/or phospholipid biosynthesis [27]. CTOss-resistance of voriconazole with other azole

antifungals has been demonstrated, probably due to common modes of action [28]. Due to the

potential for cross-resistance, specific organism susceptibility data should be reviewed before

selecting an antifungal for the treatment of infections.









Drug Stability

The stability of voriconazole has been tested under a variety of conditions. The

voriconazole chemical and physical in-use stability has been demonstrated for 24 hours at 20 to

80C (360 to 460F) following reconstitution of the lyophile with water 14]. Based on a shelf-life of

90% residual potency, voriconazole in 5% dextrose solutions were stable for at least 15 days at

40C [29]. It was reported that the stability of voriconazole in serums and plasma was found to be

stable for up to 7 days at room temperature, for 30 days frozen at -20 OC, and through 3 freeze-

thaw cycles 130]. Voriconazole diluted in 0.9% Sodium Chloride or Lactated Ringers solution can

be stable for 14 days at 150 to 300C [4]

Animal Pharmacokinetic Studies

Voriconazole's pharmacokinetics and metabolism have been studied in mouse, rat, rabbit,

dog, guinea pig, and humans after single and multiple administration by both oral and

intravenous routeS [31-36]. Its pharmacokinetics parameters have been shown in Table 1-1. Oral

absorption of voriconazole is high (~75%) in animals, including rats, mice, rabbits, guinea pigs,

and dogs [31, 37]. Drug distribution studies using radiolabeled 14C-VOriconazole 10 mg/kg in male

and female rats showed extensive distribution throughout tissues. In preclinical studies in rats,

voriconazole was excreted in both urine and feces, with the maj ority of drug eliminated within

48 hours of administration [37]

Human Pharmacokinetic Studies

Voriconazole is orally and parenterally active. Bioavailability is up to 96% with peak

plasma levels occurring 1-2 hours after dosing. Plasma protein binding is roughly 58% and is not

affected by renal or hepatic disease. The volume of distribution is 4.6L/kg, indicating widespread

distribution in the body 137, 38]. With administration of the recommended intravenous (IV) or oral

loading dose, steady-state concentrations are reached within 24 hours. Without the loading dose,









accumulation must occur for 6 days to reach steady state. Oral steady state plasma concentrations

have ranged from 2.1 to 4.8 mg/L (peak) and 1.4 to 1.8 mg/L (trough). Mean Cmax and area under

the curve (AUC) are reduced by 34% and 24%, respectively, following a high fat meal. The

absorption of voriconazole is not affected by increases in gastric pH [37, 38]. Voriconazole displays

nonlinear pharmacokinetics due to saturation of its metabolism. Increasing the IV dose from 3

mg/kg to 4 mg/kg twice daily and the oral dose from 200 mg to 300 mg twice daily results in

roughly a 2.5 fold increase in the AUC [39, 40]. Mean plasma voriconazole concentration-time

profiles after i.v. (day 7) and oral (day 14) administration are illustrated in Figure 1-2. Cmax

occurred at the end of the 1h i. v. infusion and between 1.4 and 1.8 h after oral administration [39]

Human data indicate voriconazole concentrations in the CSF are between 40% and 70% of

the concentrations in the plasma [37, 38]. Voriconazole is a substrate for, is extensively

metabolized by, and is an inhibitor of cytochrome P450 enzymes 2C19, 2C9 and 3A4. Enzyme

CYP2C 19 exhibits genetic polymorphism resulting in an approximately 4-fold higher

voriconazole exposure in poor metabolizers vs. extensive metabolizers. Eight metabolites have

been identified, of which three are maj or N-oxide metabolites. The N-oxide metabolites do not

exhibit antifungal activity. Voriconazole metabolites are primarily excreted really. Roughly

85% of a dose appears in the urine with < 2% as unchanged drug. The elimination half life has

been reported as 6 hours, but in patients receiving prolonged therapy up to 6 days have been

required to recover 90% of the drug in the urine and feceS [37, 38]. An overview of

pharmacokinetic properties of voriconazole and other triazoles from clinical studies is provided

in Table 1-2.

Inz Vitro and Inz Vivo Pharmacodynamic Studies

Pharmacodynamics refers to the time course and intensity of drug effects on the organism,

whether human or experimental animal. To describe antifungal efficacy, parameters such as









minimum inhibitory concentration (MIC), time above the MIC, time-kill curves, sub-MIC effects

and post-antifungal effects (PAFEs) are used. In vitro and in vivo studies are conducted to

examine the effects of the drug on antimicrobial inhibition or killing, the rate and extent of

killing over time, and the duration of antimicrobial effects.

Voriconazole has broad in vitro antifungal activity against a variety of fungi including

Candida spp., Aspergillus spp., Cryptococcus neoformans, Blastomyces dermatitidis,

Coccidioides immitis, Histoplasma capsulatum, Fusarium spp., and Penicillium marneffei [41-45]

The voriconazole pharmacodynamics was evaluated using time-kill methods. It reported that

voriconazole had nonconcentration-dependent activity in vitro with maximum effect observed at

3 times the MIC [46]. Voriconazole is a fungistatic agent against Candida spp. and Cryptococcus

neoformans but a fungicidal drug against Aspergillus spp. [44]. Voriconazole has shown

fungicidal activity in vitro against multiple Aspergillus spp., including A. terreus which is

inherently amphotericin B- resi stant 151. Although voriconazole MIG s for fluconazole-resi stant

isolates were significantly higher than those for fluconazole-susceptible isolates, it enhances

activity against fluconazole-resistant Candida krusei, Candida glabrata, and Candida

guilliermondii [5 6 47]. Some isolates which are resistant to fluconazole and/or itraconazole may,

expectedly, exhibit cross-resistance to voriconazole [48, 49]. Voriconazole has no activity against

the agents of Zygomycetes, such as Mucor spp. and Rhizomucor spp. which generate

considerably high voriconazole MI~s 16, 50, 51]

Voriconazole has demonstrated anti-fungi activity in a number of animal models of

systemic Candidiasis, pulmonary, disseminated and intravascular fungal infection [5' 6' 52, 53]

Moreover, it was found that serum voriconazole concentrations were very low and often

undetectable in mice, which was due to a combination of high clearance and extensive









metabolism by cytochrome P450 enzymes. Thus, mice were abandoned as being suitable for

further study of voriconazole and most subsequent work with this drug has been performed in the

guinea pig [33]. In an early study, it has been shown that in neutropenic guinea pigs with systemic

Candidiasis, voriconazole was more active than fluconazole or itraconazole in animals infected

with C. krusei, C. glabrata, or azole-resistant strains of C. albicanS [6]. A study has also been

shown that voriconazole is highly efficacious in both the prevention and treatment of Aspergillus

endocarditis in the guinea pig and is superior to itraconazole in these respects [54]. In 2000,

voriconazole was evaluated in an immunosuppressed-guinea pig model of invasive Aspergillosis.

The study indicated that voriconazole was more effective than amphotericin B or similar doses of

itraconazole and improved survival and significantly reduced tissue colony counts 1551. Recently,

some investigators have evaluated the efficacy of voriconazole in a systemic infection by

Scedosporium apiospermum in immunodepressed guinea pigs and documented that voriconazole

prolonged survival and reduced fungal load in kidney and brain tissues of the animals infected

with the first strain but was unable to prolong survival or to reduce fungal load in brain tissue for

the latter strain [56]. Voriconazole also demonstrated activity against Aspergillosis in the rat and

rabbit animal models, although the pharmacokinetics of voriconazole in these animals are

suboptimal [6]. Moreover, the Drosophila fly was also developed as a fast, high-throughput model

to study the drug efficacy against Aspergillosis and Aspergillus [57]. The study documented that

Toll-deficient Drosophila flies treated by voriconazole, had significantly better survival rates and

lower tissue fungal burdens than control.

Clinical Efficacy

Voriconazole has been proven as a promising agent for the treatment of a number of

invasive fungal infections including Candidiasis, acute and chronic invasive aspergillosis,

coccidioidomycosis, cryptococcosis, fusariosis, scedosporiosis and pseudallescheriasis,









paecilomycosis and other endemic mycoses in clinical studieS [5, 13, 58-63]. In a randomized,

double-blind, double-dummy, multicenter trial of voriconazole and fluconazole in the treatment

of esophageal candidiasis in immunocompromised patients, it proved that voriconazole was as

effective as fluconazole in the treatment of biopsy-proven esophageal candidiasis in

immunocompromised patients [64]. More recently, a study have shown that voriconazole was as

effective as the regimen of amphotericin B followed by fluconazole in the treatment of

candidemia in non-neutropenic patients, and with fewer toxic effects [65]. Interestingly,

voriconazole has shown good response against disseminated hepatosplenic aspergillosis in a

patient with relapsed leukemia, whereas itraconazole, amphotericin B and liposomal

amphotericin have no efficacy 166]. In Phase II/III trials, voriconazole was well-tolerated and

demonstrated extremely clinical efficacy in patients with fluconazole-sensitive and -resistant

candida infection, aspergillosis, and various refractory fungal infectionS [67]. The efficacy of the

combination of voriconazole and other anti-fungi drug such as caspofungin, have also studied in

the therapy for invasive aspergillosis in organ transplant. It documented that combination of

voriconazole and caspofungin might be considered preferable therapy for subsets of organ

transplant recipients with invasive aspergillosis, such as those with renal failure or A. fumigatus

infection [68]

Microdialysis

In many cases the clinical outcome of therapy needs to be determined by the drug

concentration in the tissue compartment in which the pharmacological effect occurs rather than

in the plasma. Microdialysis (MD) is an in vivo technique which allows direct measurement of

unbound tissue concentrations and permits monitoring of the biochemical and physiological

effects of drugs on the body throughout the body. MD is a catheter-based sampling technique

that provides the opportunity to sample analytes from interstitial fluid from tissues to measure









the free pharmacologically active concentration of exogenous or endogenous compounds at a site

closer to the target site than plasma [28]. The method has been used extensively in animal and

human studies for decadeS 128, 69-72]

Principle of Microdialysis

Microdialysis is a sampling technique that is used to measure the concentration of the

unbound fraction of endogenous and/or exogenous substances in the extracellular fluid of many

tissues (e.g. adipose tissue, brain, heart, lung or solid tumors) [28, 73-79]. It is applicable to both

animal and human studies. The basic principle is to mimic the function of a capillary blood

vessel by perfusing a thin dialysis probe with physiological fluid after it is inserted in the tissue

of interest by means of a guide cannula. Continuous transfer of soluble molecules from the

extracellular fluid into the probe occurs by means of a semipermeable membrane covering the tip

of the probe. Samples are subsequently collected either at intermittent time points for later

analysis by standard chemical analytical techniques, or more recently, continuously for direct

on-line analysis [28, 80-83]

Features of Microdialysis

There are many advantages of microdialysiS 128, 80-83]. It is a1 straight-forward technique that

is not too difficult to establish on a routine basis. Microdialysis sampling does not change the net

fluid balance of the surrounding sample matrix and provides clean samples in which analytes are

separated from further enzymatic action. Because there is no net fluid loss, samples can be

collected continuously for hours or days from a single freely moving animal. Most importantly,

each animal can serve as its own control, reducing the number of required subj ects compared

with some other methods such as direct tissue assay. It can be used in humans in a relatively non-

invasive manner. Moreover, microdialysis is easily coupled with other chemical analysis, such as

high performance liquid chromatography (HPLC) 183, 84], maSs spectrometry (MS) 184, 85], capillary









electrophoresis (CE) 186, 87], Or nuclear magnetic resonance (NMR) 188, 89]. These techniques are

often combined, e.g., HPLC-MS 184, 90], HPLC-NMR 189, 91], HPLC-NMR-MS 192-94] and CE-LIF

[86] to enhance the specificity, sensitivity, reliability and efficiency of separation and detection.

As with any technique, there are limitations in the application of microdialysiS 180-83, 95-97]

Microdialysis requires sensitive analytical methods to measure low concentrations in small

sample volumes. Implantation of the probe almost certainly leads to tissue reactions that can

interfere with the physiological system under investigation. To minimize this interference,

determination of optimal times after probe implantation must be determined specifically for the

analyte of interest. For example, the optimal time to measure endogenous dopamine levels after

probe insertion may not be the most optimal time to measure glutamate levels due to the

continuing process of gliosis at the site of probe tissue damage. Another problem is associated

with highly lipophilic drugs sticking to tubing and probe components, thereby complicating the

relationship between dialysate and extracellular concentrations. Most importantly, microdialysis

causes dilution of analyte levels in two ways. First, endogenous analyte levels may be decreased

near the probe leading to a tissue reaction and change in physiological status. Second, the

diluting effect of the dialysis procedure leads to lower concentrations of analyte in samples

compared to tissue, requiring both sensitive analytical methods as well as the need to determine

in vivo recovery of the analyte to calculate true concentrations in the extracellular fluid. This

latter technique may impose temporal limitations (e.g. the no-net-flux method may require hours

of sampling) and may be confounded by physiological conditions that change over time.

Application in PK/PD Studies

MD has opened many possibilities to study PK and PD processes in different tissueS 174, 98,

99]. Its many features make this technique interesting to be applied in PK and PD studies. MD

provides a clean sample, which is protein free. Because of the membrane porosity and cut-off,









large molecules such as proteins cannot get through the membrane, which allows determining the

free drug concentration in the tissue 1100, 101]. It is very important when only the free drug

concentration is considered to be pharmacologically active.

The PK/PD Approaches

Pharmacokinetics (PK) describes and predicts the time course of drug concentrations and

pharmacodynamics (PD) refers to the time course and intensity of drug effects [102-105]. PK/PD

models integrate the PK and PD approach, in which the variable of time is incorporated into the

relationship of effect to concentration [8, 102' 106]. For voriconazole, integrating the PK and PD

approaches provides information about the relationships between in vitro susceptibility, dosage,

drug concentrations in the body and antifungal or toxicological effects, which helps developers

select a rational dosage regimen for confirmatory clinical testing [106-109]

PK/PD issues have been studied most extensively for fluconazole [109-112]. A study

documented that the PD characteristic of fluconazole most closely associated with outcome was

the ratio of the area under the concentration-time curve from 0 to 24 h to the MIC (24 h

AUC:MIC) in a murine model of candidemia [109]. In a neutropenic murine model of

disseminated Candida albicans infection, the PK/PD parameters for voriconazole, ravuconazole

and posaconazole were also similarly characterized, which support that the 24h AUC/MIC ratio

is the critical PK/PD parameter associated with treatment efficacy 132, 113, 114]

However, an optimal mathematical model of voriconazole' s antifungal effect has not been

established yet. It is important to establish a mathematical model to predict clinical outcome of

voriconazole combining PK data from in vivo model studies and PD data from in vitro model

studies.

In conclusion, the studies of PK and PD profie of voriconazole play an important role in

its clinical use. Simulation through PK/PD modeling is also an important tool to capture the









variability and uncertainty that is implicitly inherent in the azole antifungal drug's further

development.

Hypothesis and Objectives

The overall hypothesis is that an optimal mathematical model of voriconazole' s antifungal

effect could be established to predict clinical outcome of them combining pharmacokinetic data

from in vivo model studies and pharmacodynamics data from in vitro model for more rational

use to improve the effectiveness of the drug therapeutics in clinic. To test the hypothesis of this

study, the following specific aims were purposed:

Specific Aim 1: Measure voriconazole activity against Can2dida isolates using time-kill

methods validated by high performance liquid chromatography.

Specific Aim 2: Apply pharmacokinetic/pharmacodynamic mathematical model to

accurately describe the activity of voriconazole against candida spp. in vitro.

Specific Aim 3: Conduct pharmacodynamic studies in dynamic system for describing the

activity of voriconazole against Can2dida spp.

Specific aim 4: Determine the value of protein binding of voriconazole in both rat and

human plasma using in vitro microdialysis.

Specific aim 5: Investigate the voriconazole PK profile from analyzing the total

voriconazole concentrations in plasma and free concentrations in muscle in rats by in vivo

microdialysis. Develop a population PK model of voriconazole in rats using NONMEM.










Table 1-1. Pharmacokinetic parameters of voriconazole in mouse, rat, rabbit, guinea pig and dog
following single and multiple administration by oral and intravenous routeS [31]
Parameter Mouse Rat Rabbit Guinea pig Dog
Sex Male Male Female Female Femalea Male/Female
Number of animals 3 2 2 3 1" 4
Plasma protein binding(%) 67 66 66 60 45 51
Intravenous
Dose (mg/kg) 10 10 10 3 10 3
Single dose AUGt (pcg-h/ml) 41.7 18.6 81.6 1.1 38.5 32.1
Multiple dose AUGt (pcg-h/ml~d 8.0 6.7 13.9 1.6 22 17.9
Oral
Dose (mg/kg) 30 30 30 10 10 6
Single dose Cmax(pcg/ml) 12.4 9.5 16.7 1 4.1 6.5
Single dose Tmax (pcg-h/ml) 2 6 1 1 8 3
Single dose AUGt (pcg-h/ml) 98.8 90 215.6 3.2 29 88.8
Multiple dose AUGt (pcg-h/ml~d 35.3 32.3 57.4 4.4 32.3 52.2
Apparent bioavailability(%) 81 159 88 87 75 138

a Male animals were used for single oral dose study.
b Number of animals per time point, except for dog and rabbit studies, which involved serial
bleeding.
cn=3 per time for single oral dose study in guinea pigs.
d Once daily for up to 10 days (minimum 5 days), except guinea pig multiple oral (three times
daily).










Table 1-2. Pharmacokinetic properties of triazole antifungals '8 115118
Property Fluconazole Itraconazole Voriconazole
Bioavailability >90% 50%-75% >95%


Posaconazole
Variable (depending on
dosage regime and food)

>98%
7-25

3-6
0.2-0.5
Hepatic: glucuronidation
to inactive metabolites

15-35 hours

<1% excreted unchanged
in urine; 66% excreted
unchanged in feces


Protein binding
Volume of
di stributi on(L/kg)
Tmax (h)
CL (L/h/kg)
Metabolism


11%
0.7-0.8


99%


58%
4.6


2-4
0.014
Hepatic:
11% metabolized


4-5
0.2-0.4
Hepatic:
CYP3A4


1-2
0.2-0.5
Hepatic:
CYP2C19, 2C9, 3A4

6-24 hours (variable)

Hepatic;
<2% excreted
unchanged in urine


Elimination half- 22-31 hours
life
Elimination route 80% excreted
unchanged in urine


35-64 hours

Hepatic;
<1% excreted
unchanged in urine





ketoconazole


i-i
N~N


fluconazole


Ia N o oa a

itr~aconazole






ac





ravuconazole


posaconazole


Figure 1-1. Structural relationship among azole drugs [5, 6]


voriconazole














Intrarvenous


2 oo
t3 mgtkg (
d TOO --4 mg/kg (





42000
100



0 2 4 6 8 10
Time peardolse
* n=13 fo~r time points at 4 and 6 h


12 14


-200 mg (n=14)
3C00 mg (n=.7)
~400 mg (n=14)


0 2 4 6 8 10 12 14
Time post-dose (h)

Figure 1-2. Plasma voriconazole concentration-time profiles following i.v. dosing (day 7) and
following oral dosing (day 14) of voriconazole 139]









CHAPTER 2
MEASUREMENT OF VORICONAZOLE ACTIVITY AGAINST CANDIDA
ISOLATES USINTG TIME-KILL METHODS VALIDATED BY HIGH
PERFORMANCE LIQUID CHROMATOGRAPHY

Background

Voriconazole is a triazole agent that inhibits ergosterol synthesis by blocking the

action of 14a-demethylase. The drug is fungistatic and exhibits no postantifungal effect

(PAFE) against Candida albicans (1, 3-5, 7, 9). Time-kill and PAFE data for Candida

glabrata are limited (1, 7) and do not exist for Candida parapsilosis isolates. Moreover,

standard time-kill and PAFE methodologies, although widely used, have not been

validated for voriconazole or other antifungals by direct measurement of drug

concentrations.

Specific Aims

This study was to develop an high performance liquid chromatography (HPLC)

assay to validate the results of time-kill and PAFE experiments for voriconazole against

C. albicans reference strains (ATCC 90029 and SC5314), and C. glabrata and C.

parapsilosis bloodstream isolates (two each).

Materials and Methods

Antifungal Agents

Stock solutions of voriconazole (Pfizer, New York, N.Y.) were prepared using

sterile water. Stock solutions were separated into unit-of-use portions and stored at -800C

until used.

Test Isolates

Eight strains were studied: 2 references (C. albicans: ATCC 90029 and SC 5314)

and 6 clinical (3 C. albicans, 2 C. glabrata and 1 C. parapsilosis).









Antifungal Susceptibility Testing

MICs were determined using E-test (Figure 2-1). The inoculum was prepared from

Sabouraud glucose agar subcultures incubated at 350C for 24 h and the resulting

suspension was adjusted spectrophotometrically to a density equivalent to a 0.5

MacFarland standard at 530 nm (1.5 x 106 CFU/ml). The solidified medium was

inoculated by dipping a sterile cotton swab into the respective undiluted stock inoculum

suspension and streaking evenly in three directions over the entire surface of a 150 mm

diameter RPMI agar plate. The plate was permitted to dry for at least 15 min before the

E-test strips with antifungal were placed on the medium surface [46]

Antifungal Carryover

Prior to time-kill experiments, assessment of the effect of solubilized voriconazole

on colony count determinations. A fungal suspension was prepared with each test isolate

to yield an inoculum of approximately 5 x 103 CFU/ml. One hundred-microliter volumes

of these suspensions were added to 900C1l volumes of sterile water or sterile water plus

voriconazole at concentrations ranging from 0.0625 to 16 times the MIC. This dilution

resulted in a starting inoculum of approximately 5 x 102 CFU/ml. Immediately following

addition of the fungal inoculum to a test tube, the tube was vortexed and a 50-C1l sample

was removed and plated without dilution on potato dextrose agar plates (Remel, Lenexa,

Kans.) for determination of viable colony counts. Following 48 h of incubation at 350C,

the number of CFU was determined. Tests were conducted in quintuplicate. The mean

colony count data for each agent at each multiple of the MIC tested were compared with

the data for the control. Significant antifungal carryover was defined as a reduction in the

mean number of CFU per milliliter of >25% compared with the colony count for the

control [46]









Time-kill Experiments

Time-kill experiments were performed in duplicate at 0.25 x, l x, 4 x and 16 x MIC.

Before testing, isolates were subcultured twice on potato dextrose agar plates. Colonies

from a 24- to 48-h culture were suspended in 9 ml of sterile water and adjusted to a

0.5 McFarland turbidity standard. 100 Cll of the adjusted fungal suspension was then

added to either growth medium alone (control) or a solution of RPMI plus an appropriate

amount of voriconazole stock solution. These procedures resulted in a starting inoculum

of approximately 1 x 105 to 5 x 105 CFU/ml concentrations and a voriconazole

concentration of 0.0625 x, 0.25 x, l x, 4 x, or 16 x MIC (Figure 2-2). Test solutions were

placed on an orbital shaker and incubated with agitation at 350C. At predetermined time

points of 2, 4, 8, 12, and 24 h, 100C1l samples were obtained from each solution, serially

diluted in sterile water, and plated (100 Cll) on potato dextrose agar plates for CFU

determination. The lower limit of reproducibly quantifiable CFU according to these

methods was 50 CFU/ml [46, 119]

Postantifungal Effect (PAFE) Experiments

PAFE experiments for control testing (no drug) and testing at 0.25, 1, 4, and 16

times the MIC in duplicate tubes. After 1 hour of incubation, the cells were centrifuged at

1,400 x g for 10 min, washed three times, and then resuspended in warm RPMI medium

(9 ml) prior to reincubation. At predetermined time points of 2, 4, 8, 12, and 24 h, 100C1l

samples were obtained from each solution, serially diluted in sterile water, and plated

(100 Cll) on potato dextrose agar plates for CFU determination [119]

HPLC for Determination of Voriconazole in Stock Solution and RPMI Medium

An HPLC protocol for measuring voriconazole in Stock Solution and RPMI media

was developed based an existing assay method with a calibration range of 0.2-10 Clg/ml









voriconazole in human plasma [120], 5-10 Clg/ml voriconazole in guinea pig plasma [121]

The analytical column was Kromasil C18, 5 mm, 250x4.6 mm (Hichrom, Reading, UK)

with a 10x3.2 mm guard cartridge (Hichrom, Reading, UK) packed with the same

material at 250C in an Agilent 1100 Series apparatus. The mobile phase was acetonitrile-

ammonium phosphate buffer (pH 6.0; 0.04 M) (1:1 v:v) and was degassed by filtration

through a 0.45 mm nylon filter under vacuum. The flow rate was 0.8 ml/min and all

chromatography was carried out at ambient temperature (~210C). Voriconazole

concentrations were determined from peak areas detected by UV absorption at 255 nm

with a retention time of 8.2 min; the maximum sensitivity was 0.025 Cg/ml. Samples of

Stock Solution and RPMI medium were diluted with 2 volumes of acetonitrile-

ammonium phosphate buffer and centrifuged at full speed in a microcentrifuge for 10

min. The supernatants were applied to the column in 200C1l sample volumes.

Calibration Curves for Voriconazole in Stock Solution and RPMI Medium

Stock solutions of voriconazole (1 Clg/CIl) were prepared in distilled water and

diluted in RPMI medium to give 100 Clg/ ml. Standards were prepared by adding the

diluted voriconazole solution to appropriate volumes of Stock Solution and RPMI

medium to give a concentrations of 0.025, 0.1i, 0.2, 0.5, 1, 2, 4, 8 and 16 Clg/ ml.

Calibration curves were constructed by plotting the peak area of voriconazole against

concentration using a weighted (1:X2) least squares regression for HPLC data analysis.

Statistical Analysis

When data are expressed as the mean + SD, group mean differences were

ascertained with analysis of variance (ANOVA). The results were considered significant

if the probability of error was < 0.05.









Results

Stability of Voriconazole in Stock Solution and RPMI Medium

The in vitro stability of voriconazole in stock solution (PH = 7.0) and RPMI

medium (PH= 7.0) were studied. There was no color change or precipitation in the

preparations and pH remained stable during the period of the studies. The results showed

that voriconazole in stock solution were found stable at -800C, -40C at least for 6 months

and one week, respectively. In RPMI medium, voriconazole were found stable at room

temperature and 370C for at least 72 hours, and stable -800C at least for 6 months (Table

2-1). The loss of voriconazole was less than 5% of the starting concentration at all

conditions.

Minimum Inhibitory Concentration (MIC)

The MICs of all isolates were within the susceptible range, as measured by Etest

and microdilution methods (Table 2-2) [122, 123]

Time-kills and Postantifungal Effect (PAFE)

For time-kills and PAFEs, colonies from 48-hour cultures on Sabouraud dextrose

agar (SDA) were suspended in 9 ml sterile water 146, 119]. One microliter of a 0.5

McFarland suspension was added to 10 ml of RPMI 1640 medium with or without

voriconazole (0.25 x, l x, 4x, and 16 x MIC), and the solution was incubated at 350C with

agitation. The maximal voriconazole concentration in these experiments was 3.04Clg/ml

(16 x MIC for C. glabrata isolate 1). For time-kills, 100 Cll from each solution was

serially diluted at desired time points (0, 2, 4, 8, 12, 24, 36, 48, 60, and 72 h) and plated

on Sabouraud dextrose agar for colony enumeration. For PAFEs experiments, Can2dida

cells were collected after 1 h of incubation, washed three times, and resuspended in warm

RPMI 1640 medium (9 ml); colonies were enumerated at the desired time points.









Voriconazole exhibited dose-response effects against all Can2dida isolates during time-

kill experiments (Figure 2-3; Table 2-2), as higher concentrations resulted in greater

growth inhibition or killing. The range of maximal growth inhibition of isolates at 1 x

and 4x MICs was -0.61- to -2.78-log and -0.53- to -2.9910g, respectively, compared to

those of controls. At 16x MIC, the range of maximal growth inhibition was -0.58- to -

4.151og (Table 2-3, 2-4, 2-5, 2-6, 2-7, 2-8 and 2-9). The results of PAFE experiments for

the tested isolates are listed in Tables of 2-10, 2-1 1, 2-12, 2-13, and 2-14. Voriconazole

did not demonstrate PAFEs as shown from Figure 2-4 which is the profile of the

concentration of colony forming units versus the time.

Voriconazole at 16 x MIC was fungicidal against C. parapsilosis isolate 2, reducing

the starting inoculum by -2.211og at 24 h (fungicidal is defined as a > 2-log reduction of

starting inoculum). Although kills did not achieve fungicidal levels for other isolates,

voriconazole at 4x and 16x MIC reduced the starting inocula of C. glabrata isolate 2 and

C. parapsilosis isolate 1 (Table 2-2). Of note, voriconazole was consistently fungistatic at

1 x to 16 x MICs, and the effect persisted for 72 h. Indeed, maximal inhibition of the four

C. albicans and C. glabrata isolates at 4x and 16x MIC (compared to starting inocula)

was not evident until 48 to 72 h. The two C. parpsilosis isolates, on the other hand,

were maximally inhibited by 24 to 36 h.

The time-kill curves of the C. parapsilosis isolates also differed from those of the

other isolates at early time points. The C. parapsilosis isolates at 4 x and 16 x MIC were

inhibited from entering the exponential growth phase, and dose-response effects were

clearly evident by 8 h. The growth rates of C. albicans and C. glabrata isolates in the

presence of voriconazole did not differ from those of controls during early exponential










phase, but dose-response effects became increasingly apparent as exponential growth

continued (8 to 24 h).

In our HPLC protocol for measuring voriconazole concentrations during time-kill

and PAFE experiments, a 250- by 4.6-mm analytic column with 10- by 3.2-mm guard

cartridge (Hichrom, Reading, United Kingdom) was packed with a 5-Clm-particle-size

Kromasil column at 250C in an Agilent 1100 series apparatus [120, 121]. Mobile phase

acetonitrile-ammonium phosphate buffer (pH 6.0; 0.04 M; 1:1 vol:vol) was degassed by

filtration through a 0.45-Clm nylon filter under vacuum; the flow rate was 0.8 ml/min.

Voriconazole concentrations were determined for peak areas detected by UV absorption

at 255 nm with an 8.2-min retention time (Figure 2-5). For each isolate, we tested RPMI

1640 medium containing at least one dose of voriconazole between lx and 16x MICs.

Samples were diluted with 2 volumes of acetonitrile-ammonium phosphate buffer and

centrifuged at full speed in a microcentrifuge for 10 min, and the supernatants (200 CIl)

were applied to the column. The maximum sensitivity was 0.025 Cg/ml, and the method

produced linear results over a range of 0.025 to 12.8Clg/ml (r2 > 0.9996). In each

instance, we confirmed that voriconazole concentrations remained constant throughout

the duration of time-kill experiments (Figure 2-6), and the drug was fully removed during

PAFE experiments (Figure 2-7).

Discussion

Voriconazole were stable at all the conditions covered in our studies. Our findings

in time-kill studies conclusively demonstrate that voriconazole exerts prolonged

fungistatic activity against C. albicans, C. glabrata and C. parapsilosis at concentrations

that are achievable in human sera with routine dosing (the medians of the average and

maximum voriconazole plasma concentrations in clinical trials were 2.51 and 3.79 Clg/ml,










respectively) [124]. Our findings are potentially relevant clinically, since certain C.

parapsilosis isolates exhibit diminished susceptibility to echinocandin antifungals and C.

glabrata isolates can develop resistance to fluconazole and other antifungal agents.

Although voriconazole caused >2-log kill of one C. parapsilosis isolate, further studies

will be needed to accurately define the extent to which the drug might be fungicidal

against clinical isolates.

To our knowledge, this is the first study to verify standard time-kill and PAFE

methodologies by directly measuring drug concentrations. We describe a simple and

reproducible HPLC method that has a broad, clinically relevant dynamic range and does

not require internal standards. The sensitivity of voriconazole measurements within

liquid media was greater than that previously reported for human or guinea pig plasma

(0.2 to 10 and 5 tol0 Clg/ml, respectively) [120, 121]. Based on our findings, we can assume

that previous studies of azoles that showed fungistatic anticandidal activity and no PAFEs

were conducted under the conditions of steady-state drug concentrations assumed by

investigators. This demonstration is crucial as efforts to use pharmacodynamic data to

develop optimal antifungal treatment strategies move forward. In particular, HPLC

methods will be essential to the design of dynamic in vitro models to assess the

pharmacodynamics of voriconazole and other agents prior to the achievement of steady-

state conditions.









Table 2-1. Percentage (%) of voriconazole maintenance in stock solution and RPMI
medium at different temperatures (Mean & SD, n = 3)
VOR in stock solution VOR in RPMI medium
Storage time 1.02 mg/ml 9.96 mg/ml 0.20 Clg/ml 1.02 Clg/ml 10.13 Clg/ml
1mon (-80 oC) 99.8f0.8 99.410.5 99.010.6 100.110.9 100.110.6
6mon (-80 oC) 98.411.7 98.810.6 97.511.7 97.411.0 99.510.6
Id (4 OC) 99.810.3
7d (4 OC) 98.610.7
24h (25 OC) 99.210.7 99.010.6 100.310.4
72h (25 OC) 99.010.7 98.611.7 98.411.9
24h (37 OC) 100.010.4 99.010.6 100.310.6
72h (37 OC) 97.010.9 97.411.6 97.710.7


Table 2-2. Minimum inhibitory concentration (MIC) of voriconazole against Candida
isolates (n=3)


C. albicans ATCC90029 0.008
C.albicans SC5314 0.012
C. glabrata: 1 0.19
C. glabrata: 2 0.032
C. parapsilosis 1 0.008
C. parapsilosis 2 0.016


Isolate


MIC(pLg/ml)














Control
1.00E+05
1.23E+05
2.84E+05
6.70E+05
1.49E+06
9.54E+06
1.07E+07


SD
0.00E+00
5.13E+03
9.81E+03
9.24E+04
2.71E+05
2.57E+05
5.78E+05


SD
0.00E+00
6.20E+03
1.28E+04
1.14E+05
1.52E+05
1.19E+06
2.00E+05


lxMIC
1.00E+05
1.18E+05
2.51E+05
5.38E+05
8.84E+05
6.33E+05
3.25E+05


SD 4xMIC
0.00E+00 1.00E+05
3.76E+03 1.22E+05
3.50E+04 2.09E+05
4.33E+04 4.87E+05
4.75E+04 6.11E+05
1.81E+04 5.58E+05
1.17E+05 2.52E+05


SD
0.00E+00
1.21E+04
3.44E+04
5.72E+04
9.06E+04
7.55E+04
8.03E+04


16xMIC
1.00E+05
1.07E+05
1.76E+05
3.85E+05
4.78E+05
5.00E+05
2.27E+05


SD
0.00E+00
2.21E+04
5.22E+04
4.89E+04
2.23E+04
6.81E+04
1.14E+05


I


Table 2-4. Voriconazole against Candida albicans SC5314 in constant concentration experiments (Data of CFU/mL, Mean & SD, n
2)
Time (h) Control SD 0.25xMIC SD 1 xMIC SD 4xMIC SD 16xMIC SD
0 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00
2 9.67E+04 1.71E+03 9.46E+04 1.51E+04 1.14E+05 8.02E+03 9.28E+04 1.15E+04 8.20E+04 9.21E+03
4 1.77E+05 2.11E+04 1.28E+05 6.96E+03 1.28E+05 2.66E+04 1.18E+05 1.97E+04 1.00E+05 2.11E+04
8 8.78E+05 2.44E+05 5.31E+05 1.92E+05 3.99E+05 1.35E+05 2.75E+05 8.31E+04 2.88E+05 1.10E+05
12 1.08E+06 1.96E+05 7.14E+05 4.46E+04 4.24E+05 1.70E+05 2.97E+05 3.88E+04 3.54E+05 1.53E+05
24 1.30E+06 1.06E+05 7.93E+05 5.74E+04 2.80E+05 6.57E+04 3.40E+05 1.40E+05 2.95E+05 1.13E+05
48 1.25E+06 1.55E+05 7.00E+05 9.44E+04 2.46E+05 9.56E+04 2.79E+05 8.14E+02 2.71E+05 6.26E+04


Table 2-3. Voriconazole against Candida albicans ATCC90029 in constant concentration experiments (Data of CFU/mL, Mean & SD,
n = 3)


0.25 xMIC
1.00E+05
1.04E+05
2.06E+05
5.43E+05
9.38E+05
3.32E+06
2.54E+06


Time (h)
0
2
4
8
12
24
48











Table 2-5. Voriconazole against Candida glabrata 1 in constant concentration experiments (Data of CFU/mL, Mean & SD, n = 2)
Time (h) Control SD 0.25 xMIC SD l xMIC SD 4xMIC SD 16xMIC SD
0 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00
2 1.06E+05 1.77E+04 1.03E+05 1.24E+04 1.08E+05 1.58E+04 1.05E+05 2.04E+04 1.00E+05 1.80E+04
4 1.15E+05 1.47E+04 1.11E+05 2.79E+04 1.46E+05 6.29E+04 1.19E+05 3.41E+04 1.31E+05 3.37E+04
8 1.76E+05 7.00E+04 1.54E+05 6.41E+04 1.63E+05 7.19E+04 1.61E+05 5.35E+04 1.70E+05 5.59E+04
12 5.25E+05 6.45E+03 5.35E+05 4.15E+04 4.22E+05 1.07E+05 2.85E+05 2.55E+04 3.61E+05 2.85E+04
24 2.46E+06 5.70E+04 1.39E+06 2.76E+05 7.32E+05 1.80E+05 6.96E+05 1.52E+05 6.20E+05 1.10E+04
48 3.86E+06 2.49E+05 2.19E+06 3.90E+04 1.05E+06 4.87E+05 7.45E+05 8.02E+04 4.36E+05 4.11E+04








Table 2-6. Voriconazole against Candida glabrata 2 in constant concentration experiments (Data of CFU/mL, Mean & SD, n = 2)
Time (h) Control SD 0.25 xMIC SD l xMIC SD 4xMIC SD 16xMIC SD
0 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00
2 9.60E+04 3.97E+02 1.01E+05 1.63E+04 1.01E+05 3.62E+03 8.67E+04 1.41E+04 9.96E+04 1.83E+04
4 1.20E+05 2.64E+03 1.25E+05 2.84E+04 1.24E+05 5.68E+03 1.11E+05 1.26E+04 1.11E+05 2.68E+04
8 4.86E+05 2.51E+05 4.28E+05 1.66E+05 3.52E+05 7.74E+04 2.78E+05 8.54E+04 2.81E+05 9.87E+04
12 1.23E+06 4.34E+05 8.90E+05 6.26E+04 6.03E+05 1.25E+05 3.37E+05 4.88E+04 2.94E+05 4.36E+04
24 5.34E+06 9.61E+05 1.08E+06 2.01E+05 5.80E+05 2.78E+04 4.26E+05 1.39E+04 3.52E+05 6.21E+04
48 7.11E+06 1.41E+06 2.96E+05 1.42E+05 6.91E+04 3.16E+04 3.74E+04 1.26E+04 3.47E+04 1.44E+04











Table 2-7. Voriconazole against Candida parappisiloi 1 in constant concentration experiments (Data of CFU/mL, Mean & SD, n


Time (h)
0
2
4
8
12


Control
1.00E+05
1.06E+05
1.67E+05
9.97E+05
1.84E+06


SD
0.00E+00
8.26E+03
6.03E+03
9.36E+03
4.52E+05


0.25 xMIC
1.00E+05
9.41E+04
1.41E+05
6. 39E+0 5
1.64E+06


SD
0.00E+00
2.25E+04
1 .3 3E+04
1.98E+04
3.35E+05


lxMIC
1.00E+05
9.62E+04
1.39E+05
2.71E+05
3.74E+05


SD
0.00E+00
8.26E+03
2.05E+02
5.42E+04
2.73E+04


4xMIC
1.00E+05
1.00E+05
1 36E+0 5
2. 30E+0 5
1.48E+05


SD
0.00E+00
6.56E+03
3.57E+03
3.12E+03
7.51E+04


16xMIC
1.00E+05
1.03E+05
1.3 0E+05
1.94E+05
1.25E+05


SD
0.00E+00
2.44E+03
8.87E+03
3.28E+03
4. 19E+04


7.26E+06 2.44E+06 5.32E+06
3.45E+07 2.51E+06 2.72E+07


1.51E+06 6.67E+05 1.41E+05 1.25E+05 8.30E+03 1.21E+05 1.34E+04
1.06E+07 3.47E+06 5.73E+05 1.28E+05 3.71E+04 9.06E+04 3.60E+04


Table 2-8. Voriconazole against Candida parappisiloi 2 in constant concentration experiments (Data of CFU/mL, Mean & SD, n


Time (h)
0
2
4
8
12
24
48


Control
1.00E+05
1.10E+05
1.36E+05
3.51E+05
1.16E+06


SD
0.00E+00
7.07E+03
1.70E+04
1.77E+04
2.05E+05


0.25 xMIC
1.00E+05
1 .04E+0 5
1.25E+05
4.25E+05
9. 15E+05


SD
0.00E+00
1.41E+03
2. 12E+03
7.85E+04
9.69E+04


lxMIC
1.00E+05
9.49E+04
1.09E+05
7.44E+04
8.09E+04


SD
0.00E+00
3.18E+03
1.27E+04
1.58E+04
2.00E+04


4xMIC
1.00E+05
9.55E+04
1.08E+05
6.47E+04
3.07E+04


SD
0.00E+00
2.33E+04
9.90E+03
1.88E+04
9.76E+03


16xMIC
1.00E+05
8.55E+04
8.00E+04
2.32E+04
1.26E+04


SD
0.00E+00
7.00E+03
2.68E+04
7.79E+03
1.41E+03


5.60E+06 3.04E+05 6.08E+06 2.57E+06 7.12E+04 3.04E+03 1.15E+04 2.98E+03 2.78E+03 3.07E+03
1.02E+07 2.21E+06 1.42E+07 4.53E+06 5.35E+05 5.69E+05 5.15E+04 2.87E+04 1.73E+03 7.14E+02





































SD
1.56E+04
2.05E+04
1.48E+05
7. 64E+0 5


lxMIC
2.95E+05
2.89E+05
5.15E+05
1.66E+06


SD
7.78E+03
2. 12E+03
3.54E+04
2.47E+05
7.07E+04
1.06E+06
4.88E+06


4xMIC
3.15E+05
3.28E+05
6.20E+05
1.91E+06
4.35E+06
1.60E+07
2.24E+07


SD
7.07E+03
5.66E+03
1.27E+05
3.11E+05
2. 12E+05
4. 17E+06
1.34E+06


16xMIC
2.87E+05
3.35E+05
5.75E+05
2. 17E+06
5.3 0E+06
1.61E+07
2.93E+07


SD
1.48E+04
6.36E+04
1 34E+0 5
1.20E+06
7.07E+05
3.68E+06
5.02E+06


2.83E+05 5.40E+06 4.24E+05 4.65E+06
1.27E+06 1.80E+07 3.89E+06 1.46E+07
2.69E+06 2.84E+07 7.07E+06 2.77E+07


Table 2-9. Summarized voriconazole time-kill data against Candida isolates
Maximum log kill at":
Isolate MIC(Clg/ml)b 24h 24-48h >48h
lx MIC 4x MIC 16x MIC lx MIC 4x MIC 16x MIC lx MIC 4x MIC 16x MIC
C. albicans ATCC90029 0.008 -1.23 -1.34 -1.34 -1.77 -1.84 -1.96 -1.63 -1.90 -1.89
C.albicans SC5314 0.012 -0.61 -0.51 -0. 54 -0. 56 -0.53 -0.58
C. glabmata 1 0. 19 -0.43 -0.49 -0.59 -0. 74 -0.78 -1.00 -0.99 -1.08 -1.17
C. glabmata 2 0.032 -1.02 -1.10 -1.15 -2.23 -2.39 -2.49 -2.78 -2.99 -3.02
C. parapsilosis 1 0.008 -0.99 -1.64 -1.68 -0. 91 -2.67 -2.74 -0. 67 -2.37 -2. 69
C. parapsilosis 2 0.016 -1.86 -2.70 -3.93 -1.03 -2. 13 -3.94 -0.88 -1.12 -4.15
Time-kill data are presented as the maximum difference between the growth of the control isolate (grown in the absence of voriconazole) and that
of the isolate in the presence of various voriconazole concentrations at < 24 h, 24 to 48 h, and > 48 h. The maximum growth inhibition of each
isolate at the given concentrations is shown in bold.
SMICs, as determined by Etest and the Clinical and Laboratory Standards Institute (formerly NCCLS) broth microdilution method ['3, were
within twofold agreement.


Table 2-10. Voriconazole against Candida albicans ATCC90029 in PAFE experiments (Data of CFU/mL, Mean & SD, n =2)


Time (h)
0
2
4
8
12
24
48


Control
3.00E+05
3.20E+05
5.65E+05
1.75E+06
5.10E+06
1.55E+07
2.56E+07


SD
1.98E+04
2.83E+03
7.07E+03
1.41E+05


0.25xMIC
3 .04E+0 5
3.15E+05
6. 15E+05
1.99E+06

































Control
2.58E+05
2.3 7E+0 5
3.40E+05
8.60E+05
1.41E+06
7.80E+06
1.15E+07


0.25 xMIC
2.35E+05
2. 15E+05
3.07E+05
6.70E+05
1.62E+06
7.00E+06
1.29E+07


lxMIC
2.58E+05
2.3 7E+0 5
3.53E+05
1.06E+06
1.53E+06
7.00E+06
1.35E+07


4xMIC
2.67E+05
2. 39E+0 5
2.60E+05
7.00E+05
1.05E+06
7.80E+06
1.16E+07


16xMIC
2.77E+05
2.64E+0 5
3 10E+0 5
1.03E+06
1.15E+06
6.80E+06
1.34E+07


Table 2-11i. Voriconazole against Candida glbrata 2 in PAFE experiments (Data of CFU/mL, Mean & SD, n =2)
Time (h) Control SD 0.25xMIC SD l xMIC SD 4xMIC SD 16xMIC SD
0 2.61E+05 1.77E+04 2.31E+05 2.12E+03 2.38E+05 4.24E+03 2.62E+05 4.95E+03 2.44E+05 1.27E+04
2 2.48E+05 2.47E+04 2.18E+05 4.81E+04 2.66E+05 1.48E+04 2.43E+05 5.23E+04 2.30E+05 1.77E+04
4 5.55E+05 2.12E+04 4.80E+05 4.24E+04 5.20E+05 4.24E+04 6.00E+05 2.83E+04 5.30E+05 2.83E+04
8 1.37E+06 7.07E+04 1.48E+06 1.34E+05 1.20E+06 7.07E+03 1.58E+06 9.90E+04 1.79E+06 1.20E+05
12 3.19E+06 6.36E+04 3.08E+06 8.49E+04 3.35E+06 2.12E+05 3.00E+06 1.41E+05 3.60E+06 1.06E+05
24 1.48E+07 7.07E+05 1.59E+07 2.76E+06 1.80E+07 4.81E+06 1.54E+07 6.36E+05 1.67E+07 4.24E+05
48 1.91E+07 5.66E+05 1.54E+07 1.41E+05 1.75E+07 1.34E+06 1.72E+07 3.54E+05 1.97E+07 1.56E+06


Table 2-12. Voriconazole against Candida glabrata 1 in PAFE experiments (Data of CFU/mL)


Time (h)
0
2
4
8
12
24
48





Table 2-13. Voriconazole against Candida parappisiloi 1 in PAFE
CFU/mL)


experiments (Data of


Time (h)

2
4
8
12
24
48


Control
2.89E+05
2.51E+05
4.80E+05
9.90E+05
2.40E+06
1.46E+07
3.22E+07


0.25 x1VIC
3.40E+05
2.85E+05
6.80E+05
1.69E+06
3.42E+06
2. 18E+07
3.04E+07


1 x1VIC
3 10E+0 5
2.95E+05
6.40E+05
1.12E+06
2.50E+06
1.47E+07
3.02E+07


4 x1VIC
2.70E+05
2.46E+05
6.00E+05
1.55E+06
3.04E+06
1.64E+07
3.03E+07


16 x1VIC
2.84E+05
2.67E+05
6.27E+05
1.70E+06
3.16E+06
2.00E+07
3.69E+07


Table 2-14. Voriconazole against
CFU/mL)


Candida parappisiloi 2 in PAFE experiments (Data of


Time (h) Control 0.25 x1VIC
0 1.81E+05 1.79E+05


1 x1VIC
1.42E+05
1.40E+05
1.54E+05
3.66E+05
1.33E+06
1.45E+07
2.36E+07


4 x1VIC 16 x1VIC
1.55E+05 1.79E+05
1.46E+05 1.34E+05
2.20E+05 1.40E+05
6.50E+05 4.70E+05
1.92E+06 1.48E+06
1.95E+07 1.21E+07
2.43E+07 2.00E+07


1.38E+05
1 34E+0 5
4. 10E+0 5
1.34E+06
1.37E+07
2.26E+07


1.36E+05
1.61E+05
6.00E+05
1.40E+06
1.60E+07
2.29E+07






















A B
Figure 2-1. Susceptibility of voriconazole against Candida isolates were determined by
disc diffusion, E-test and the Clinical and Laboratory Standards Institute broth
microdilution method. A) Disc diffusion. B) E-test.


Figure 2-2. Culture flasks for voriconazole against Candida isolates in vitro
















-*- ctrl C. albicans sc5314
-o- 0.25*MIC
-9- 1*MIC
-A- 4*MIC
-m 16*MIC


le+7l l*c -m- 16*Mic-- -- le+7 -




l1e+6 -( bY L le+6



le+5 l+




le+4 .. 1+4
0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr


le+8 -*- ctrlC. glabrata 1 l+
-0 0.25*MIC

-6 4*MIC le+8
le+7 -- 1*MIC


E~ le+7

u..1e+6

z' 'p'I z le+6

le+5
le+5


1e44 le+4
0 4 8 12 16 20 24 28 32 36 40 44 4E
Time hr


0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr


0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72
Time hr


le+9



le+8



Sle+7


Z le+6



le+5


le+4


le+8



le+7



~le+6



z'le+5



le+4



le+3


0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr Time_hr

Figure 2-3. Time-kill curves for voriconazole against Candida isolates (Mean + SD, n=3

for C. albicans ATCC90029, n=2 for other tested strains, without significant

differences in results by ANOVA)








































0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr


l e+8





le+7
E


z
le+6





le+5





l e+8





l e+7





l e+6





l e+5


C. glabrata 2



1ele



le+7




-a 0.25*MIC
1*MIC
-A 4*MIC
-N- 16*MIC
le+5
0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr


0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr


0 4 8 12 16 20 24 28 32 36 40 44 48 0 4 8 12 16 20 24 28 32 36 40 44 48
Time hrTieh


Figure 2-4. Voriconazole did not demonstrate PAFEs. (Mean + SD, n=2 for C. albicans

ATCC90029, C. glabrata 2, without significant differences in results by

ANOVA; n=1 for other three tested strains.)







DAD14 d9 ig255,4 Ref-360,100 (lVY5JO7\011-1101 D)


:FC


Figure 2-5. Chromatogram of voriconazole which concentration was determined from
peak area detected by UV absorption at 255 nm with around an 8.2 minute
retention time


I j


0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr
Figure 2-6. Voriconazole concentrations in culture media throughout the duration of
time-kill experiments by HPLC: Representative data for a 48-h time-kill
experiment with C. glabrata isolate 1 are presented. (Mean a SD, n=3,
without significant differences in results by ANOVA)
















40-



C 20-




0 -10-
O


o


f -30-

-40-

-5 0 I I I I I I I I I I I *
0 4 8 12 16 20 24 28 32 36 40 44 48
Time hr

Figure 2-7. Voriconazole concentrations in PAFE experiments by HPLC: Representative
data for a 48-h time-kill experiment with C. glabrata isolate 1 are presented.
(Mean + SD, n=3, without significant differences in results by ANOVA)









CHAPTER 3
APPLYING PHARMAC OKINETIC/PHARMAC OD YNAMIC MATHEMATIC AL
MODEL ACCURATELY DESCRIBES THE ACTIVITY OF VORICONAZOLE
AGAINST CANDIDA SPP. IN VITRO

Background

Time-kill curves are attractive tools for studying the pharmacodynamics of

antimicrobial agents as they provide detailed information about antimicrobial efficacy as

a function of both time and concentration [125]. Although time-kill curves can be studied

using animal models of infection, in vitro models offer significant advantages in cost,

convenience and time, as well as permitting direct investigation of the antimicrobial-

microbe interaction in a controlled and reproducible manner [126]. The results of specific

time-kill experiments can be accurately described using pharmacokinetic/

pharmacodynamic (PK/PD) mathematical models. By using PK data derived from

clinical trials or animal experiments, mathematical models can be used to simulate the

expected kill curves for different doses and dosing regimens of an antimicrobial agent

[127]. As such, PK/PD modelling is a potentially powerful technique for defining optimal

antimicrobial treatment strategies.

Voriconazole is a triazole antifungal agent effective in the treatment of patients

with candidiasiS [8, 65]. We recently performed time-kill experiments against two strains

each of Can2dida albicans, Candida parappisiloi and Can2dida glabrata using

voriconazole concentrations of 0.25 x, lx, 4x and 16x the minimum inhibitory

concentration (MIC) and measuring colony counts at 0, 2, 4, 8, 12, 24, 36, 48, 60 and 72

h [22]. We demonstrated that voriconazole exhibited dose-response effects against all

isolates, as higher concentrations resulted in enhanced killing of Can2dida spp. The ranges

of maximal growth inhibition of isolates at concentrations of lx and 4x MIC were -0.61









to -2.78 log and -0.53 to -2.99 log, respectively, compared with controls. At 16 x MIC,

the range of maximal growth inhibition was -0.58 to -4. 15 log.

Specific Aim

The obj ectives of the present study were to develop a PK/PD mathematical model

to fit our time-kill data and to use the model to simulate the expected kill curves for

typical intravenous and oral dosing regimens.

Materials and Methods

Mathematical Modelling of Time-kill Data

The methods and results of antifungal susceptibility testing and time-kill curve

experiments for voriconazole against Candida isolates were described previously [22]

During time-kill experiments, aliquots were taken at each time point and examined under

the microscope to exclude hyphal formation or clumping. The voriconazole MICs against

the isolates ranged from 0.008 to 0. 19Clg/mL. Fitting of time-kill data was performed

with a series of models using the Scientist@ 3.0 non-linear least squares regression

software program (MicroMath, Salt Lake City, UT), as detailed in the Results and

Discussion section. To determine the model that best fits the data for each isolate, graphs

were visually inspected for goodness of fit and several criteria were considered, including

model selection criterion (MSC), coefficient of determination (R2) and the correlation

between measured and calculated data points.

Simulations of Expected Time-kill Curves Using Human PK Data

PK parameters were extracted from published data sets of humans receiving

voriconazole using XyExtract v2.5(Wilton and Cleide Pereira da Silva, Campina Grande,

Para'lba, Brazil) and the data were analyzed by WinNonlin 5.2 (Pharsight Corporation,

Mountain View, CA) to describe the best absorption model. The Scietntist' 3.0 software










program was then used to simulate plasma concentration-time profiles for multiple

dosing of voriconazole (z = 12 h). Simulations of expected kill curves for each isolate

were made using steady-state PK parameters in the mathematical model that was shown

earlier to best fit the time-kill data.

Statistical Analysis

When data are expressed as the mean + SD, group mean differences were

ascertained with analysis of variance (ANOVA). The results were considered significant

if the probability of error was < 0.05.

Results

An Adapted Sigmoidal Emax Model Provides the Best Fit for Voriconazole Time-kill
Data against Candida Isolates

Fitting of the time-kill data was started with a simple, commonly used Emax model

[127].


= k max N(31
s~t= ECo +Ck C1 31

In this model, dN/dt is the change in number of Can2dida as a function of time, ks (h ) is

the candidal growth rate constant in the absence of voriconazole, kmax (h ) is the

maximum killing rate constant (i.e. maximum effect),ECso (mg/L) is the concentration of

voriconazole necessary to produce 50% of maximum effect, C (mg/L) is the

concentration of voriconazole at any time (t) and N (colony-forming units (CFU)/mL) is

the number of viable Can2dida. Since voriconazole concentrations did not change during

the time-kill experiments, C was constant for the entire fitted time period.

The Emax model does not account for the fact that isolates have not yet reached the

logarithmic growth phase at time zero or for the delay in the onset of killing by









voriconazole. To compensate for these delays, an exponential correction factor (1-exp-zr

was incorporated into the preceding model. Since the delays in the onset of candidal

growth and voriconazole killing were not necessarily equivalent, they were assigned

different adaptation rate constants (i.e. delay in growth = (1-exp-ar and delay in killing =

(1-exp '))>:

dN k C
= k, 1 e-at max -1 32
dt ( -ea ECo +C -ePi 32


Model (2) was adapted by considering the maximum number of Can2dida (Nmax),

which potentially corrected for the limitations of space and nutrients that are inherent in

in vitro systems:


= k 1 1 -e-at max -1 e ) 3 3
dt mxECo +Cj 3)

In the final model, a Hill factor (h) was incorporated, which modified the steepness

of the slopes and smoothed the curves:


dN lI N k, Ch
dtNm -a EC7, +Ch e"r 34

The fitted time-kill curves of the six Can2dida isolates were generated using the

different models in the absence of voriconazole (control) or in the presence of constant

concentrations (mg/L) of voriconazole. In Model (4), Hill factor values ranging from 0 to

4 were tested. In general, the curve fittings for all PD data were acceptable. The best fits

for each isolate were obtained using Model (4); the PD parameters and goodness of fit

criteria were excellent for all voriconazole-Ca~ndida isolate pairings, with MSC>2.46 and

R2 >0.95. The fitted curves for Model (4) are shown in Figure 3-2. The PD parameters of









voriconazole against the six Can2dida isolates and the goodness of fit criteria are

presented in Table 3-1.

In time-kill experiments, PD effects can be described by at least three parameters:

ks, the growth rate constant of the organism in the absence of drug; kmax, the maximum

killing rate constant of the drug against the organism; and EC50 (the concentration of

drug necessary for 50% of its maximum effect) [125, 127]. The ECso values calculated in the

present study demonstrate that voriconazole was highly effective at easily achievable

serum levels against each of the six Candida isolates (ECso range 0.002-0.05 mg/L)

(Table 3-1). Moreover, the rates of maximal killing by voriconazole were highly

correlated with the growth rates of the isolates (Pearson' s correlation coefficient =

0.9861). In other words, the data suggest that voriconazole achieved its highest kill rates

against the most rapidly proliferating organisms.

Human PK Data for Voriconazole Can Be Used to Simulate Expected Time-kill
Curves

Mathematical models can be used to simulate the expected kill curves for different

doses and dosing regimens of antimicrobials by substituting the concentration term with

PK parameters collected in vivo and accounting for protein binding

[127]. To simulate time-kill curves that might be expected for voriconazole under

typical dosing regimens, we first extracted PK parameters from published data sets of

humans receiving the drug intravenously (3 mg/kg twice a day (bid)) and orally (200 mg

bid) in steady-state 139]. Plasma concentration profiles for intravenous and oral dosing

were found to be best characterized by two-compartment models of zero- and first-order

absorption, respectively. Some of the PK parameters are listed in Table 3-2.










Applying the PK parameters for the two dosing regimens, we simulated plasma

concentration-time profiles for multiple dosing of voriconazole (r = 12 h), assuming

plasma protein binding of 58% (Figure 3-2) 181. We then used the steady-state

PK parameters to generate simulations for voriconazole activity against each of the

Can2dida isolates in the present study (Figure 3-2). As shown, voriconazole showed

fungistatic activity against all six strains. For five of the six strains, simulated kills were

virtually identical for the intravenous and oral regimens. The exception was C.

parapsilosis 2, for which the effects of intravenous dosing were greater. By collecting PK

data for a range of voriconazole doses and dosing regimens during clinical trials or

animal models, we should be able to use the mathematical model to predict treatment

strategies that might be most effective against a given Can2dida isolate.

Discussion

A pharmacokinetic/pharmacodynamic (PK/PD) mathematical model was also

developed in our studies to fit voriconazole time-kill data against Candida isolates in

vitro and used the model to simulate the expected kill curves for typical intravenous and

oral dosing regimens. Although PD studies comparing the efficacy of various

voriconazole regimens against Candida isolates are feasible in animal models 132], they

are laborious, expensive and complicated.1In vitro models permit direct study of the

interaction between fungi and fungal agents in a controlled and reproducible manner and

allow direct comparisons of different agents and dosing strategies in a more convenient

and faster way [126]

We have shown that it is feasible to describe precisely the time-kill activity of

voriconazole against Can2dida isolates by using an adapted sigmoidal Emax model that

corrects for delays in candidal growth and the onset of voriconazole activity, saturation of










the number of Can2dida that can grow in the in vitro system and the steepness of the

concentration-response curve. We established the PK/PD models and simulated the

expected voriconazole activity for typical intravenous and oral dosing regimens by using

PK data collected in vivo and accounting for protein binding. A limitation of the present

study design was that voriconazole concentrations were constant. In the future, we will

characterize voriconazole-Candida interactions using a dynamic in vitro system in which

drug concentrations change over time in a manner consistent with the PK profile in

humans. Based on the success of the present study, it should also be feasible to model

accurately the time-kill data for changing voriconazole concentrations. In conclusion, we

demonstrated that the activity of voriconazole against Can2dida isolates can be accurately

described using a mathematical model. We anticipate that this approach will be an

efficient method for devising optimal voriconazole treatment strategies against

candidiasis.















L. parapsdlosts 1 L. parapsdlosts 2
0.002/0.008/0.032/0.128 0.004/0.016/0.064/0.256
0.22 0.66
0.24 0.63
0.0066 0.012
0.26 0.072
0.23 0.19
3.48E+07 6.95E+06
1.49 3.84
5.63/0.997 4.91/0.95


ks: fungal growth rate constant in the absence of voriconazole
kmax: maximum killing rate constant (maximum effect)
ECso: concentration of voriconazole necessary to produce 50% of maximum effect
alpha: constant used to fit the initial lag phase for the growth
beta: constant used to fit the initial lag phase for the inhibition or killing
h: Hill factor
MSC: model selection criteria


__ __


Table 3-1. Pharmacodynamic parameters and goodness of fit criteria against Candida isolates
Parameter(unit) L. aD2cants Al LL90029 L. abscants SLb33 i L. gzabrata 1 L. g8baa2
C (mg L) 0. 002/0. 008/0. 032/0. 128 0.003/0.012/0. 048/0. 192 0.0475/0. 19/0.76/3.04 0. 008/0. 032/0. 12 8/0.512
K, (h ') 0.28 0.87 0.24 0.26
Kma ( )0.34 0.72 0.29 0.35
Elns. s0.0015 0.0030 0.05 0.0042
alpha 1.24 0.19 0.14 0.58
beta 0.069 0.11 0.032 0.059
Nmax(CFU mL) 1.09E+07 1.21E+06 3.85E+06 6.57E+06
h 2.13 2.77 1.16 2.32
10SC/R^2 4.68/0.99 3.58/0.99 4.76/0.995 4.33/0.99









Table 3-2. Steady-state pharmacokinetic parameters in plasma as calculated by two-
compartment model analysis
Dose regimen
Parameter (unit) 200mg oral 3mg/kg/h i.v .
Cmax (mg/L) 1.59 2.66
Tmax (h) 1.72 1.00

ac (h- ) 0.443 1.260

P (h- ') 0. 139 0. 120
F 0.9 1.0
CL/F (L/h) 17.9 14.2
Vdss/F (L) 98.9 109.7
Vdarea/F (L) 128.5 117.9
i.v. : intravenous
Cmax: maximum concentration
Tmax: time to Cmax
a: macro rate constant associated with the distribution phase
fl: macro rate constant associated with the elimination phase
F: bioavailability
CL/F: systemic clearance
Vdss/F: apparent volume of distribution at steady-state
VdareaF: apparent volume of distribution











Irtment i.v. short-term infusion 3r


5000


E **** VORIlfree
S4000-



Z 3000-



4 2000-








0 10 20 30 40 50 60

Time post-dose (h)


3000 2 compartment 200mg oral
VORI total

E 2500- **** VORI_free


.0 2000


g 1500


E1000 ,, *
C ** *

S 500 i*,** *



0 10 20 30 40 50 60
Time post-dose (h)


Figure 3-1. Plasma VOR concentration-time profiles simulated with a two-compartment
PK model. (PK data were extracted from published paperl39]). Short-term
intravenous (3mg/kg/h, bid, T=1h); Oral administration (200mg, bid, F=0.9);
fb = 58%

















1018 C. albicans ATCC90029
SN1 contml
O N20002mglL
I N30008mglL
A N4 0032mglL
SN50O128mglL
le+7 N1 cal
**** N2_cal















0 12 24 36 41
Time hr

C. glabrata 1


lso _


10.8 C. albicans SC5314
SN1:control
O N2:o.oo2mgil
7 N3:0.008mgil
& N4:0.032mgil
H N5:0.128mgil
3 le+7 N cal
***N2 csl


E
u
O
Z


12 24

C. glabrata 2


SN1 contml
O N200475mgil
y N3019mglL
& N4076mglL
SN53j04mglL
- N1 cal
**** N2 cal

-- N3 ca


12 24 36 48


12 24 36 48 60 72


012 24 36 48 0 12 24 36 48


Figure 3-2. Fitted time-kill curves derived using the mathematical model for constant

concentrations of voriconazole (Mean + SD; n = 3 for Can2dida albicans

ATCC90029 and n = 2 for the other strains, without significant differences in

results by analysis of variance (ANOVA)). Voriconazole concentrations were

O x, 0.25 x, l x, 4 x and 16 x the minimum inhibitory concentrations (listed as

N1-N5, respectively, in the individual figure legends).
















































le+4

le+3

le+2



le-2






-le+4

-le+3


I"- le+1

-le+0


C. albicans ATCC90029


le+8

le+7

le+6

le+5


l +3


le+1

le+0


le-1

l e-2


le+8

le+7

le+6

le+5



l e+3 ~


le+1

le+0

le-1

l e-2


10 20 30
Time hr

C. arlabrata 1


40 50 60


C. albicans SC5314le
le8

le+7

le6


le+5






le+1



le-1

l e-2
0 10 20 30 40 50 60


C. glabrata 2


1


0 10 20 3 40 0 6


C. parapsilosis 2


/C~
....

....

.~~~~.







Cc


010 20 30 40 50 60 0 10 20 30 40 50 60


Figure 3-3. Simulations of candidal time-kills and plasma voriconazole concentration-

time profiles. The simulations oftime-kills are shown for short-term

intravenous (i.v.) infusion (----) and oral dosing of voriconazole (---). Control

growth is in the absence of drug (-). The simulated plasma concentration-

time profiles for multiple doses of voriconazole (z = 12 h) are shown for

short-term i.v. infusion (z = 1 h, infusion rate = 3 mg/kg/h, twice a day (bid))

and oral administration (200 mg, bid, F = 0.9). Free voriconazole

concentrations were calculated assuming plasma protein binding level of 58%.









CHAPTER 4
PHARMACODYNAMIC STUDY IN DYNAMIC SYSTEM FOR DESCRIBING THE
ACTIVITY OF VORICONAZOLE AGAINST CANDIDA SPP. IN VITRO

Background

Voriconazole is a lipophilic triazole antifungal agent that exhibits potent activity in vitro

and in vivo against a variety of fungi including Can2dida spp., Aspergillus spp., Cryptococcus

neoformans, Bla~stomyces dermatitidis, Coccidioides immitis, Histopla~sma capsulatum,

Fusarium spp. and Penicillium marneffei [41-43]. Using standard in vitro time-kill methodologies,

we and others have shown that voriconazole exerts prolonged fungistatic activity and no post-

antifungal effect against diverse Can2dida spp., including C. albicans, C. glabrata and C.

parapsilosis [8, 22]. In a follow-up study, we developed a mathematical model that accurately

described the activity of voriconazole against Can2dida spp. during time-kill experimentS [102]

After identifying and validating the in vitro pharmacodynamic (PD) parameters derived from our

experiments, we used pharmacokinetic (PK) data extracted from human data sets to simulate the

expected kill curves in vivo for typical intravenous and oral dosing regimens of voriconazole.

Indeed, by using PK data from human clinical trials or animal studies to simulate kill curves for

different doses and dosing regimens of voriconazole or other antimicrobial agents, PK/PD

modeling is potentially a powerful technique for defining optimal treatment strategies [127]

The maj or limitation of our earlier studies was that voriconazole concentrations remained

constant over time at Eixed multiples of the minimum inhibitory concentration (MIC) [22, 46, 102, 128,

129]


Specific Aim

The first obj ective of the present study was to develop a dynamic time-kill methodology in

which voriconazole concentrations diminished in a manner consistent with the serum PK profile

in humans. After we characterized the interaction between voriconazole and Hyve Can2dida










isolates using the dynamic in vitro system, our obj ectives were to fit our mathematical model to

the changing concentration kill-curves and to use human PK data to simulate expected kill curves

mn vivo.

Materials and Methods

Antifungal Agents

Stock solutions of voriconazole (Pfizer, New York, N.Y.) were prepared using sterile

water. Stock solutions were separated into unit-of-use portions and stored at -800C until used.

Test Isolates

Five strains were studied: 1 reference (Can2dida albicans ATCC 90029) and 4 clinical (2

Candida glabrata and 2 Candida parapsilosis).

Softwares

PK parameters were extracted from published data sets of humans receiving voriconazole

using XyExtract v2.5(Wilton and Cleide Pereira da Silva, Campina Grande, Para'lba, Brazil) and

the data were analyzed by WinNonlin 5.2 (Pharsight Corporation, Mountain View, CA) to

describe the best absorption model. The Scietntist' 3.0 software program was then used to

simulate plasma concentration-time profiles for multiple dosing of voriconazole (z = 12 h).

Dynamic Model Design

In the infection model, the voriconazole was introduced as an intravenous bolus and was

eliminated at constant rate according to the first order kinetics:

C = Co e-ket (4-1)


C = (4-2)









Where Co is the initial voriconazole concentration in the flask, t is the time that has elapsed since

the loading of voriconazole, C is the voriconazole concentration at time t, and ke is the

voriconazole elimination rate constant.

For any two points of time t,

C', = Co, -et (4-3)


Cr 2 0I k t 2 (4-4)


C, =~ (4-5)


A, Al C, -V
C 2_ 2 (4-6)
Vo Vo

From equations 4-3, 4-4, 4-5 and 4-6, equation 4-7 could be derived:


Co e-ket, 0, -et 0 (4-7)


Reconstitute equation 4-7,

AV = Vo -(1- e-k(ty-t2 ) (4-8)

As At = ti- t2,

AV = Vo -(1 e-ke' ") (4-9)

Where Vo is the initial volume of medium in the flask, and At is the time period of replacing

voriconazole-free medium.

The Ke could be calculated using a certain half-life (tl/2):

In 2
Ke= (4-10)









As shown in Figure 4-1, every 2 hours (At), 4.13 ml (AV) of voriconazole containing

medium were replaced by the same volume of voriconazole-free medium in order to simulate an

in vivo half-life of 6 hours.

Time-kill Experiments in the Dynamic Model

The MICs of voriconazole against a C. albicans reference strain (ATCC 90029) and C.

glabrata and C. parapsilosis clinical isolates (two each), as determined by E-test and the Clinical

and Laboratory Standards Institute (formerly NCCLS) broth microdilution method, were

consistent with previous reports [22, 46, 123], ValUeS were within two-fold agreement by the two

methods (Table 2-1). In our previous time-kill experiments, voriconazole at constant

concentrations of l x, 4x and 16xMIC inhibited growth of the isolates in a dose-dependent

manner [22]. The dynamic time-kill experiments were performed in duplicate against each isolate

(inocula of 1 x105 to 5 x105 CFU/mL). Colonies from 24-hour cultures on Sabouraud dextrose

agar (SDA) plates were suspended in 9 mL of sterile water and adjusted to UV absorbance of

0.84-0.86 at 53 5nm; 200C1L of the suspension were added to 20 mL (Vo) of RPMI 1640 (Sigma

Chemical Co.) buffered to a pH of 7.0 with MOPS (morpholinepropanesulfonic acid; RPMI

1640 medium) without voriconazole, and incubated at 350C with continuous agitation. After 2

hours pre-incubation, voriconazole was added at starting concentrations of 0.25 x, lx, 4x, or

16xMIC. The drug was introduced as a bolus and eliminated at constant rate according to first

order kinetics which was described previously. Prior to the removal of voriconazole-containing

medium at each time point, 100 CIL aliquots were obtained from each solution, serially diluted,

and plated on SDA plates to enumerate CFUs. Voriconazole concentrations during time-kill

experiments were verified by serial HPLC measurements [22]









Results

Voriconazole exhibited dose-response effects against all Candida isolates during the

dynamic time-kill experiments, as higher starting concentrations resulted in greater growth

inhibition or killing (Figures 4-2). The time-kill data in the dynamic time-kill experiments for

individual isolates were listed in the tables of Table 4-1, 4-2, 4-3, 4-4, 4-5. Of note, starting

concentrations of 4x and 16xMIC resulted in growth inhibition or killing of all five isolates;

results with 16xMIC starting concentrations were clearly superior to 4xMIC starting

concentrations for four isolates (C. albicans ATCC90029, C. glabrata 2, C. parapsilosis 1, and

C. parapsilosis 2). On balance, the results were consistent with those observed for constant

concentrations of 4x and 16xMIC against the same isolates in our earlier time-kill study [22] I

the present study, results with lx MIC starting concentrations were significantly inferior to 4x

and 16xMIC against all five isolates (Figure 4-2). In fact, lx MIC starting concentrations caused

slight growth inhibition of only two strains (C. albicans ATCC90029 and C. glabrata 1); for the

remaining three isolates, the 1 x MIC growth curves did not differ significantly from controls

grown in the absence of voriconazole. These results differed from constant concentrations of l x

MIC in the previous study, for which growth inhibition and kills of all isolates were significantly

compared to controls. In fact, growth inhibition and kills at constant concentrations of 1 x MIC

approached those seen at constant concentrations of 4x and 16xMIC against three isolates (C.

albicans ATCC90029, C. glabrata 1 and C. glabrata 2).

We used our previously developed sigmoidal Emax-model to fit time-kill data for each

voriconazole-Candida pairing in the dynamic experimentS [102, 127]. The model accounts for

delays in the onset of both Candida growth and voriconazole-induced kill, the saturation of

Candida growth, and incorporates a Hill factor that modifies the steepness of the slopes and

smoothes the curves:










dN N ,k Ch
dt l Ni Ni EC jh +h I(-1
maxx 50


In this model, dN/dt is the change in number of Can2dida as a function of time, ks (h ) is

the candidal growth rate constant in the absence of voriconazole; kmax (h ) is the maximum

killing rate constant; ECso (mg/L) is the concentration of voriconazole necessary to produce 50%

of maximum effect; C (mg/L) is the concentration of voriconazole at any time (t); N (CFU/mL)

is the number of viable Can2dida; a is the constant used to fit the initial lag phase for the growth;

p is the constant used to fit the initial lag phase for the inhibition or killing; and h is the Hill

factor.

Simultaneous fitting of the model to the changing concentration kill-curves was performed

using Scientist3.0 non-linear least squares regression software program (MicroMath, Salt Lake

City, UT) as previously described [102]. Graphs were visually inspected for quality of fit and

several criteria were considered, including model selection criterion (MSC), coefficient of

determination (R2) and the correlation between measured and calculated data points. For each

isolate, goodness of fit criteria was excellent, with MSC > 2.71 and R2 > 0.97 (Table 4-6, Figure

4-3). Similarly, PD parameters were excellent (Table 4-6). The ECso values (drug

concentrations necessary for 50% of maximum effect) indicate that voriconazole was highly

effective against each of the five isolates at easily achievable serum levels (range: 0.005 to 0.08

mg/L); these values were comparable to those obtained in the earlier constant concentration

experiments (range: 0.0015 to 0.05 mg/L). As in the previous study, voriconazole achieved its

highest rates of kill against the most rapidly proliferating isolates.

To simulate the expected kill curves for different doses and dosing regimens of

voriconazole, we substituted the concentration term in our model with PK parameters collected









in vivO [39, 102]. PK data were extracted from published data sets of humans receiving

voriconazole intravenously (3 mg/kg twice a day (bid)) or orally (200 mg bid) using XyExtract

v2.5 (Wilton and Cleide Pereira da Silva, Campina Grande, Paraiba, Brazil). Plasma

concentration profiles for intravenous and oral dosing were best characterized by two-

compartment models of zero- and first-order absorption, respectively (analysis by WinNonlin

5.2, Pharsight Corporation, Mountain View, CA). PK parameters for the two dosing regimens

are presented in Table 2. These values were used to simulate the respective voriconazole plasma

concentration-time profiles (z = 12 h; Scietntist' 3.0 software program) (Figure 4-4). The PK

parameters for two dosing regimens were then applied to the Emax-model to simulate the

expected in vivo kill curves, assuming plasma protein binding of 58% 181

Simultaneous fitting of the identified model to the changing concentration kill-curves was

performed in Scientist@ 3.0 non-linear least squares regression software program (MicroMath,

Salt Lake City, UT) as previously described [102]. To determine the model that best fits the data

for each isolate, graphs were visually inspected for quality of fit and several criteria were

considered, such as model selection criterion (MSC), coefficient of determination (R2) and the

correlation between measured and calculated data points.

In the current study, we employed a dynamic in vitro model to determine time-kill curves

of voriconazole against C. albicans reference strain, C. glabrata and C. parapsilosis bloodstream

isolates (2 each) at x0.25, xl, x4 and 16xMIC. The MICs were determined by Etest and the

Clinical and Laboratory Standards Institute (formerly NCCLS) broth microdilution method [22, 46,

123], and were within twofold agreement (Table 2-1).

We characterized voriconazole-Candida interactions using the dynamic in vitro system in

which drug concentrations change over time in a manner consistent with the PK profile in









humans [39, 102] (Table 3-2). PK parameters were extracted from published data sets of human

receiving voriconazole using XyExtract v2.5 (Wilton and Cleide Pereira da Silva, Campina

Grande, Paraiba, Brazil) and the data were analyzed by WinNonlin 5.2 (Pharsight Corporation,

Mountain View, CA) to describe the best absorption model. The Scientist@ 3.0 software

program was then used to simulate plasma concentration-time profiles for multiple dosing of

voriconazole (z = 12 h). Simulations of expected kill curves for each isolate were made using

the PK parameters for two dosing regimens in the Emax-model as described above, assuming

plasma protein binding of 58% 181

As shown in Figure 4-4, the simulated kill curves predicted that voriconazole would exert

prolonged fungistatic activity against all five isolates, resulting in diminishing Candidal

concentrations over time. Indeed, for three isolates (C. albicans ATCC90029, C. parapsilosis 1

and C. parapsilosis 2), reductions from starting inocula over 60 hours were predicted to exceed

2-logs. Of note, kill curves for four isolates were almost identical for the IV and oral dosing

regimens. The exception was C. parapsilosis 2, for which the effects of IV dosing were

predicated to be greater than oral dosing.

Discussion

Our study is noteworthy for several reasons. To our knowledge, this is the first report of

time-kill experiments against Candida isolates using an in vitro method that mimics changing

voriconazole concentrations in vivo rather than testing constant concentrations. The most

striking finding from our study was that starting voriconazole concentrations of 1 x MIC were

largely ineffective at inhibiting the growth of isolates, whereas starting concentrations of both 4x

and 16xMIC significantly inhibited all isolates. The data from the dynamic model, therefore,

suggest that clinicians should aim for peak serum concentrations of voriconazole > 4xMIC in

treating their patients with candidemia. Since the median of maximum voriconazole plasma









concentrations in clinical trials was 3.79 Clg/ml [124], a target peak concentration > 4xMIC fits

well with the recently proposed breakpoint MICs < 1 Clg/mL for voriconazole susceptibility [122]

Second, we demonstrated that it is possible to accurately fit the data from our dynamic

model using an adapted Emax mathematical model. Moreover, we used the mathematical model

to simulate expected kill curves for typical dosing regimens of voriconazole by substituting the

concentration term with PK data collected in vivo and accounting for protein binding. In this

regard, our approach of combining in vitro time-kill data with existing in vivo PK data might

serve as a model for future studies to define the optimal use of antifungal agents. Although PD

studies comparing the efficacy of various regimens of voriconazole and other antifungals are

feasible in animal models 132], they are complicated, laborious and expensive. In vitro time-kills

permit direct study of the interaction between antifungals and fungi in a controlled and

reproducible manner, and they allow direct comparisons of different agents and dosing strategies

in a more convenient, faster and cheaper way that does not expend animal liveS [126]. Finally, our

data predict that typical IV and oral dosing regimens of voriconazole are comparable in their

ability to kill most Candida isolates in vitro. As such, the data lend support to the practice of

switching to oral voriconazole to complete a course of therapy of invasive candidiasis among

patients who have exhibited clinical and microbiologic responses to initial parenteral antifungal

therapy.

In conclusion, we have shown that the PK/PD relationship of voriconazole against Candida

spp. can be accurately modeled in a dynamic in vitro system that mimics in vivo conditions.

PK/PD modeling approaches such as ours merit further investigation as tools to define optimal

treatment regimens of voriconazole and other antifungal agents against candidiasis.











Table 4-1. Voriconazole against Candida albicans ATCC90029 in changing concentration experiments (Data of CFU/mL, Mean &
SD, n = 3)
Time (h) Control SD 0.25 xMIC SD l xMIC SD 4xMIC SD 16xMIC SD
0 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00
2 2.13E+05 5.98E+04 1.66E+05 4.04E+04 1.63E+05 3.32E+04 1.85E+05 3.86E+04 1.72E+05 5.24E+04
4 3.25E+05 7.81E+04 2.72E+05 5.40E+04 2.84E+05 2.16E+04 2.50E+05 3.81E+04 2.42E+05 8.13E+04
6 4.62E+05 8.61E+04 4.25E+05 8.52E+04 3.89E+05 4.60E+04 2.92E+05 4.01E+04 2.96E+05 6.86E+04
8 6.83E+05 1.51E+05 6.25E+05 7.77E+04 4.46E+05 6.18E+04 3.30E+05 7.16E+04 1.62E+05 2.58E+03
10 1.06E+06 1.67E+05 9.69E+05 1.08E+05 8.45E+05 9.47E+04 2.00E+05 1.71E+04 7.80E+04 3.92E+03
12 1.88E+06 2.36E+05 1.94E+06 5.03E+05 9.42E+05 7.51E+04 1.27E+05 3.16E+03 3.03E+04 4.35E+03
14 2.53E+06 5.20E+05 2.67E+06 6.74E+05 1.04E+06 1.85E+04 9.23E+04 6.31E+03 2.13E+04 6.21E+03
16 4.71E+06 1.62E+06 3.66E+06 1.54E+06 1.38E+06 2.71E+05 7.99E+04 1.39E+04 1.86E+04 4.58E+03
18 7.05E+06 3.85E+06 6.03E+06 2.85E+06 1.91E+06 3.47E+05 7.24E+04 1.83E+04 1.18E+04 2.54E+03
20 9.43E+06 1.68E+06 8.21E+06 1.83E+06 2.46E+06 2.00E+05 6.83E+04 1.41E+04 1.07E+04 1.99E+03
22 1.11E+07 2.89E+06 9.34E+06 2.91E+06 3.36E+06 3.72E+05 4.04E+04 9.00E+03 9.42E+03 1.10E+03
24 1.45E+07 1.48E+06 1.15E+07 7.47E+05 3.82E+06 4.91E+05 4.42E+04 5.10E+03 1.10E+04 2.96E+03












Table 4-2. Voriconazole against Candida glabrata 1 in changing concentration experiments (Data of CFU/mL, Mean & SD, n = 2)
Time (h) Control SD 0.25 xMIC SD l xMIC SD 4xMIC SD 16xMIC SD


1.00E+05
1 .42E+0 5
2. 04E+0 5
2. 59E+0 5
3.27E+05
3.90E+05
6. 18E+05
9.20E+05
1.32E+06
1.90E+06
2.66E+06
4.03E+06
4.13E+06


0.00E+00
1.67E+04
1.58E+04
5.81E+04
4.91E+04
1 17E+0 5
2. 36E+0 5
9.91E+04
2.05E+04
2.96E+04
2.28E+05
8.75E+05
7.98E+05


1.00E+05
1 17E+0 5
1.66E+05
2.29E+05
2. 72E+0 5
2.96E+05
5 50E+0 5
1.01E+06
1.13E+06
1.48E+06
1.60E+06
2.37E+06
2.97E+06


0.00E+00
2.07E+04
1.06E+04
8.03E+03
5.58E+04
4.97E+04
1.47E+05
4.91E+04
8.66E+04
1.88E+05
2.40E+04
3.97E+05
6.44E+0 5


1.00E+05
1.29E+05
1.95E+05
2.49E+05
2.60E+05
2.92E+0 5
4.60E+05
7.20E+05
8 32E+0 5
1.06E+06
1.19E+06
1.70E+06
1.76E+06


0.00E+00
2.37E+03
2.68E+04
8.48E+04
8.17E+04
1.07E+05
1.73E+05
4.3 0E+04
3.31E+04
2.52E+04
9.47E+04
2. 11E+04
5.27E+04


1.00E+05
1.04E+05
1.64E+05
1.87E+05
2.13E+05
2.54E+05
3.59E+05
3.58E+05
5.05E+05
5.69E+05
5.57E+05
5.38E+05
5.60E+05


0.00E+00
1.04E+04
5.69E+04
5.75E+04
6.09E+04
7.58E+04
7.56E+04
1.45E+04
1.15E+05
6.91E+04
5.27E+04
3.66E+04
1.68E+04


1.00E+05
1 .04E+0 5
1.68E+05
1.88E+05
1.87E+05
2.01E+05
3 .3 4E+0 5
3.65E+05
4.40E+05
4.46E+05
4.64E+05
4.01E+05
3 .8 9E+0 5


0.00E+00
7.22E+03
4.36E+04
4.00E+04
1.67E+04
3.16E+04
7.96E+04
8.58E+04
9.53E+02
1.37E+03
4.67E+02
1.39E+04
1.40E+04











Table 4-3. Voriconazole against Candida glabrata 2 in changing concentration experiments (Data of CFU/mL, Mean & SD, n = 2)
Time (h) Control SD 0.25 x1VIC SD 1 x1VIC SD 4 x1VIC SD 16 x1VIC SD
0 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00 1.00E+05 0.00E+00
2 1.57E+05 5.31E+04 1.31E+05 1.09E+04 1.41E+05 2.86E+03 1.10E+05 1.18E+04 1.21E+05 9.25E+02
4 2.21E+05 1.58E+04 1.79E+05 1.08E+04 2.27E+05 2.20E+04 1.82E+05 7121E+03 1.69E+05 4.62E+04
6 3.07E+05 7.51E+04 2.61E+05 3.51E+04 3.79E+05 8.15E+04 2.42E+05 8.49E+04 2.07E+05 5.75E+04
8 4.35E+05 1.20E+05 3.84E+05 2.51E+04 4121E+05 9.94E+04 3120E+05 9.63E+04 2.40E+05 1.87E+04
10 5.43E+05 8.13E+04 5.41E+05 1.96E+03 5.96E+05 1.72E+05 4.04E+05 9.96E+04 3.06E+05 4.68E+04
12 9.80E+05 4.67E+05 7128E+05 7.96E+04 6.89E+05 1.70E+05 4.28E+05 7.68E+04 3.33E+05 4.29E+04
14 1.24E+06 2120E+05 1.07E+06 5129E+01 8.48E+05 2.28E+05 4.48E+05 8.55E+04 3.53E+05 3.85E+04
16 2122E+06 8.49E+05 1.81E+06 3.12E+05 1.50E+06 4.50E+05 7.17E+05 2.06E+05 4.55E+05 1.12E+04
18 3.54E+06 1121E+06 2.80E+06 1.03E+06 2.52E+06 9.34E+05 7.77E+05 1.56E+05 5.06E+05 1.37E+05
20 4.40E+06 1127E+06 3.42E+06 8.45E+05 3.11E+06 8.76E+05 8.36E+05 1.72E+05 4.93E+05 9.74E+04
22 5.80E+06 2.40E+06 4.87E+06 1.16E+06 3.30E+06 8.54E+05 7.65E+05 1.87E+05 4.45E+05 7.21E+04
24 5.95E+06 1.77E+06 5.54E+06 1.46E+06 3.42E+06 8.15E+05 7.89E+05 2.01E+05 4.32E+05 3.21E+04












Table 4-4. Voriconazole against Candida parapsilosis 1 in changing concentration experiments (Data of CFU/mL, Mean & SD, n = 2)
Time (h) Control SD 0.25 xMIC SD 1 xMIC SD 4 xMIC SD 16 xMIC SD


1.00E+05
1.86E+05
4.80E+05
8.96E+05
1.70E+06
2.43E+06
3.60E+06
5.88E+06
8.48E+06
1.06E+07
1.54E+07
1.89E+07
2.27E+07


0.00E+00
7.74E+04
2.20E+05
3 56E+0 5
4.70E+05
4.01E+05
1.38E+06
1.88E+06
2.81E+06
2.20E+06
2.45E+06
4.79E+06
3.06E+05


1.00E+05
1.89E+05
3.66E+05
6.49E+05
1.46E+06
2.3 0E+06
3.63E+06
5.35E+06
8.55E+06
1.21E+07
1.43E+07
1.27E+07
1.67E+07


0.00E+00
1.03E+04
2.91E+04
4. 19E+04
1.08E+05
4.47E+05
2.02E+06
2.58E+05
1.99E+05
2.89E+06
1.46E+06
1 17E+0 5
7. 17E+06


1.00E+05
2.06E+05
3.87E+05
6. 16E+0 5
9.61E+05
1.60E+06
3.02E+06
4.71E+06
6.44E+06
8.27E+06
1.13E+07
1.11E+07
1.37E+07


0.00E+00
4.45E+04
3.68E+04
4.00E+03
3.02E+05
3.22E+05
1.29E+06
2.37E+06
2.76E+06
2.81E+06
2.20E+06
3.23E+06
3.39E+06


1.00E+05
1.80E+05
4. 18E+05
4.96E+05
1.88E+05
1.44E+05
1.11E+05
1.37E+05
1.71E+05
1.64E+05
2. 16E+05
3.20E+05
5.32E+05


0.00E+00
4.99E+04
2.75E+04
9.34E+04
3.16E+04
6.45E+03
2.48E+04
1.23E+04
2. 10E+04
4.07E+04
9. 19E+03
8.26E+04
2. 16E+05


1.00E+05
1 .72E+0 5
2.73E+05
3.87E+05
1 .64E+0 5
7.09E+04
3.27E+04
4.42E+04
5.59E+04
4.54E+04
5.28E+04
5.73E+04
5.95E+04


0.00E+00
4.79E+04
9.65E+04
1.77E+05
7. 10E+04
1.95E+04
8.60E+03
1.74E+04
2.58E+04
2.31E+04
2. 01E+04
2.71E+04
2.56E+04












Table 4-5. Voriconazole against Candida parapsilosis 2 in changing concentration experiments (Data of CFU/mL, Mean & SD, n = 2)
Time (h) Control SD 0.25xMIC SD l xMIC SD 4xMIC SD 16xMIC SD


1.00E+05
1.85E+05
2.58E+05
3.80E+05
5.20E+05
1.62E+06
2.32E+06
2.23E+06
3.09E+06
4. 16E+06
4.57E+06
5.23E+06
5.73E+06


0.00E+00
3.82E+04
3.3 0E+04
1.86E+05
2.29E+05
4.52E+05
3.79E+04
3.06E+05
5 .82E+0 5
1.28E+06
1.12E+06
4.25E+05
1.41E+06


1.00E+05
1.63E+05
2.75E+05
3.07E+05
6. 18E+05
1.59E+06
2.51E+06
2.74E+06
2.96E+06
5.42E+06
5.23E+06
5.47E+06
6.83E+06


0.00E+00
4.84E+04
4.05E+04
1.21E+05
1.21E+05
2. 18E+04
4.20E+05
3.13E+04
5 50E+05
2.64E+06
1.25E+05
8.45E+05
1.86E+06


1.00E+05
1.35E+05
2. 37E+0 5
3.60E+05
6. 14E+0 5
1.23E+06
1.87E+06
2.97E+06
2.86E+06
4.25E+06
4. 19E+06
4.86E+06
3.79E+06


0.00E+00
4.03E+03
5.48E+04
1.76E+05
2.69E+05
3.62E+05
2.45E+05
4.3 0E+03
2. 18E+05
1.47E+06
1.05E+06
6.69E+05
4.23E+05


1.00E+05
1 .84E+0 5
1 19E+0 5
1.06E+05
1.07E+05
8.58E+04
8.59E+04
9. 11E+04
1.67E+05
2.97E+05
3 50E+0 5
4.97E+05
7.40E+05


0.00E+00
8.69E+04
3.61E+04
4.04E+04
3.22E+04
5.19E+03
4.97E+04
4.51E+04
9.81E+03
5.19E+04
5.63E+04
2.07E+05
3.68E+05


1.00E+05
1.33E+05
5.66E+04
5.40E+04
2.86E+04
2.22E+04
1.22E+04
1.17E+04
1.44E+04
1.16E+04
9.13E+03
1.63E+04
1.29E+04


0.00E+00
7.26E+04
4.88E+03
1.47E+04
3.83E+03
1.35E+04
6.37E+03
1.35E+03
2.70E+02
4.84E+03
3.80E+03
7.84E+03
6.77E+03













Table 4-6. Pharmacodynamic parameters and goodness of fit criteria against Candida isolates in the dynamic infection model
Parameter(umit) L. abscans Al 009029, L. glabrata 1 L. glabrata 2 L. parapsilosis 1 L. parapsilosis 2


C (mg/L)a
Ks (h')


ECso (mg/L)
a
(3
Nmax (CFU/mL)
h
MSC/R2


0.002/0.008/0.032/0. 128 0.0475/0. 19/0.76/3.04 0.008/0.032/0. 128/0.512 0. 002/0. 008/0. 032/0.128 0. 004/0. 016/0. 064/0. 256
0.29 0.17 0.19 0.32 0.30
0.45 0.27 0.38 0.49 0.50
0.005 0.074 0.0810 0.007 0.017
1.21 5.55 6.98 9.81 9.34
0.556 0.061 0.085 0.25 0.70
1.65E+07 1.87E+07 2.48E+07 1.96E+07 5.88E+06
1.42 0.86 0.89 1.79 2.86
3.58/0.98 3.68/0.99 2.71/0.97 3.19/0.98 3.08/0.98


Cmax: maximum voriconazole concentration
ks: fungal growth rate constant in the absence of voriconazole
kmax: maximum killing rate constant (maximum effect)
2?ECso: concentration of voriconazole necessary to produce 50% of maximum effect
a: constant used to fit the initial lag phase for the growth
p: constant used to fit the initial lag phase for the inhibition or killing
h: Hill factor
MSC: model selection criteria

a Values of voriconazole concentrations of 0.25 x, lx, 4x and 16x MIC, respectively.











C. parapsilosis 1


0.008
--0. 007
0.006
2.0.005s
oi 0.004
o 0.003
*C 0.002

0.000


y = 0.0087e-o.1144x
R2 = 0.9931


0 2 4 6 8 10 12 14 16 18 20 22 24

Time(h)


C. parapsilosis 2


0.016
0.014
0.012
0.010
0.008
0.006
0.004
0.002
0.000


y = 0.0174e-o.1141x
R2 = 0.9929


0 2 4 6 8 10 12 14
Time (h)


16 18 20 22 24


Figure 4-1. Concentration elimination curve of voriconazole in the dynamic infection model
















C albicans ATCC90029

tN1ontrol
SN20O002mglL
SN30O008mglL
SN4003O2mglL
SN50O128mglL


4 8 12 16 20 24

C. parapsilosis 1




NI~~ roto


le+8-




le+6-



2 le+5-



le+


Sle+8






le+5-



le+4-




le*o




le+7


le+6


le+5




le+4


le+3-


0 4 8 12
Time hr
C glabrata 1


16 20 24


C glabrata 2
~Nlrontrol
O N2 O 008mgll
t N9 O 092m911
-b NI O 128mgll
--)N10112mgll






_ieL--~


~Nlrontrol
~ N2 O Oll~mgll
~ N9 O 19mgll
~ NI O 16mgll
--) NI 9 Olmgll


0 4 8 12 16 20 24

le~o -C. parapsilosis 2





le+6 -( N06ml



Sle+5


le+4


0 4 12 6 20 24 0 4 8 12 16 20 24




Figure 4-2. Time-kill curves from the dynamic in vitro model (Mean a SD; n=3, without

significant differences in results by ANOVA). Voriconazole concentrations were x,

0.25 x, 1 x, 4 x and 16 xMICs (listed as N1-N5 in the individual figure legends).






















le+7 m N5 a12amgi ..*
**N2 cal *
-N3cal








1013


3 N1 coro a N c'








ZIle+4 le+4X







0 4 8 1 16 2 240 4 8 12 16 20 24

le+9C. p arapslosi 1 lsC. p larpslosi 2
SNiontrol I Nicomroln0
O NzoaimgmyL O NzomamgrL
y NI0mamyL N3ColemgrL
le+8 A N40~mgmL Ae* N40mImgrL
Ni01 cal N1 c


1e6 Nso ,~~T--~-- le+61 sus I adi:~~b"












1030 4 8 12 16 20 24 130 4 8 12 16 20 24



cocetrtin of voriconazole (Ma SD;8 n=3 without osgifan difrncsi
resulnts yAOV) oicnzl cocntrantionswr ,02 ,lx n
16MO (listedy as N1N5i te nivdulfue lgnd)













le+8
le+7
-ctri growth_CFUlmL
le+6 .I I -- oral kill_CFUlmL
***** IV kill CFUlmL
le+5 ~-, --- Ct_oral_mglL
le4 `"--.- Cf_oral_mglL
le+4 ---- Ct iv mglL
t v C ivmglL
le+3-
le+2


le+0 -PLr~;` ~ y 1
le-1
010 20 30 40 50 60
Time hr
le+8C. glabrata 1 ies C. glabrata 2















le~eC. prapiloss 1 e~eC. parapsilosis 2





le+4 le+











Figure 4-4. Using parameters from dynamic models to simulate candidal time-kills and plasma
voriconazole concentration-time profiles: The simulations oftime-kills are shown for

short-term IV infusion (----) and oral dosing of voriconazole (---). Control growth is in

the absence of drug (-). The simulated plasma concentration-time profiles for

multiple doses of voriconazole (z=12h) are shown for short-term intravenous (IV)

infusion (T=1h, infusion rate=3mg/kg/h, bid); and oral administration (200mg, bid,

F=0.9). Free voriconazole concentrations were calculated assuming plasma protein

binding level of 58%.






















80









CHAPTER 5
IN VITRO MICRODIALYSIS OF VORICONAZOLE

Background

Microdialysis is a dynamic technique which has been employed for the in vivo

measurement of a variety of drugs and endogenous compounds in different tissueS 173, 99]

The applicability of this technique can be limited by drug lipophilicity which can impair

the diffusion through microdialysis semi-permeable membrane [130, 131]. Voriconazole is a

moderately lipophilic (Log D7.4 = 1.8) antifungal triazolic agent 131]. The determination

of the recovery (the fraction of the free drug at the site of sampling which dialyses into

the probe) is essential for voriconazole to estimate the free drug levels in the tissue. There

are several factors which affect drug's relative recovery including the drug's physico-

chemical properties, flow rate, probe's characteristics such as type, membrane length and

diameter, experimental conditions such as temperature and matrix tortuosity of the

interested tissues, when the recovery is determined in vivO [99]. For determining the in

vitro recovery of voriconazole in microdialysis, a straight-forward approach was

undertaken and the calibration methods directly were used to determine the free, unbound

drug concentration. For practical reasons, the two most common methods are the

extraction efficiency method (EE) and retrodialysis (RD) method.

Specific Aims

The obj ective of in vitro microdialysis (MD) study was to assess the feasibility of

doing microdialysis of voriconazole, to determine the in vitro recovery of this compound

and to determine the value of protein binding of voriconazole in both rat and human

plasma in vitro.












Materials

All materials are listed in Table 5-1 unless otherwise stated.

The CMA/20 probe, with a membrane length of 10 mm and a molecular cutoff of

20 kDa was used for the experiments. The probe was connected to a 1000 Cl1 Gastight

syringe by a Perifix screw connector (B. Braun). A microinfusion pump (Harvard

Apparatus Syringe Pump Model '22') was used to keep the flow through the probe

constant.

All the solvents used in the HPLC analysis were of HPLC grade.

Methods

Recovery

The in vitro recovery of voriconazole was determined by two different methods:

extraction efficiency (EE) and retrodialysis (RD). All methods were carried out at 37oC.

Each procedure is described in the following sections.

Extraction efficiency method (EE)

In the EE method, blank Lactated Ringer' s (LR) solution was pumped through the

microdialysis (MD) probe at a flow rate of 1.5 CL/min. The MD probe was then placed

into the sample tube, which was a glass tube filled with approximately 6 ml of drug

solution, starting with the lowest concentration and fixed in place with adhesive tape. Six

different voriconazole concentrations of 0.5, 1, 2, 5, 10 and 40 Cpg/ml were used in this

experiment with all prepared in lactated Ringer' s solution. To guarantee equal

concentrations throughout the whole tube, the solution was stirred at approximately

100rpm. Voriconazole was diffused from the sample tube into the MD probe and were

measured in the dialysate. Samples were collected every 20 minutes after the end of the 2


Materials and Methods









hours equilibration time. A total of 5 samples were collected for each experiment. The

concentrations were determined by a validated HPLC/UV method and plotted versus

time. All experiments were performed in triplicates. The percent recovery (R%) for the

EE method is then calculated as followed:


R% = dialsaex 100 (5-1)
~sol

Where R%/ is the recovery in percentage, Csol is the average voriconazole concentration in

the tube before and after the experiment, and Cdialysate is the voriconazole concentration in

the dialysate.

Retrodialysis method (RD)

In RD method, the syringes contained the voriconazole solution that was pumped

through the MD probe at a flow rate of 1.5 CL/min. The MD probe was placed into a

sample tube that was filled with blank LR solution and fixed in place with adhesive tape.

To guarantee equal concentrations throughout the whole tube, the solution was stirred at

approximately 100rpm. The voriconazole was diffused out of the probe into the sample

tube. The loss of voriconazole over the membrane was then determined while

determining Cdzalysate. Samples are collected every 20 minutes after the end of the 2 hours

equilibration time. A total of 5 samples were collected for each experiment. The

voriconazole concentrations were determined by a validated HPLC/UV method and

plotted versus time. All experiments were performed in triplicates. The percent recovery

(R%/) for the RD method was then calculated as followed:


R% = p~uie dnye x 100 (5-2)
~perfusate










Cperfitsate is the average voriconazole concentration in the perfusate before and after

the experiment, and Cdzalysate is the voriconazole concentration in the dialysate.

Microdialysis experiments

The in vitro recovery of the voriconazole was determined by the extraction

efficiency method (EE). In this method, blank Ringer' s solution was pumped through the

microdialysis probe (CMA/20), which was placed into a testing tube filled with

approximately 6 ml of voriconazole solution. Six different voriconazole concentrations of

0.5, 1, 2, 5, 10 and 40 Cpg/ml were tested in this experiment with all prepared in lactated

Ringer' s solution. After placing the probe in the drug solution, it was allowed to

equilibrate for 10 minutes at flow rate of 5p~l/min followed by 2 hours at 1.5Cpl/min.

Subsequently dialysate samples were collected every 20 minutes after the equilibration

period. A total of 5 samples were collected for each experiment, and the experiment was

triplicate for each concentration. The voriconazole concentration in the dialysates and in

the tube before and after the experiments was determined by the HPLC/UV method

described above. The probe recovery determined by the extraction efficiency was

calculated by the equation 5-1. The protein binding was determined by the extraction

efficiency was calculated by the equation 5-3.


protein binding % = (1- dialysate ) x100 (5-3)
R% x C,,

HPLC for determination of voriconazole

An HPLC protocol for measuring voriconazole in Lactated Ringer' s solution and

human or rat plasma was developed based an existing assay method with a calibration

range of 0.025-16 Clg/ml voriconazole as being described previously in Materials and

Methods of Chapter 2. Voriconazole samples in Lactated Ringer' s solution were









measured using HPLC/UV method directly. Plasma samples were extracted before the

measurement using HPLC/UV method as being described previously [120] with

modification. Briefly, plasma samples were centrifuged at 1,700 x g for 5 min. An aliquot

(150 CIl) was pipetted into a 1.5 ml Eppendorf tube and acetonitrile (240 CIl) was then

added. The mixture was mixed briefly by vortex, centrifuged at 1,200 x g for 5 min after

standing for 10 min at room temperature and supernatant was transferred to a new

Eppendorf tube. This step was repeated one more time. Then 50 Cll of the supernatant

plus 50 Cll of 0.04 M Ammonium Phosphate buffer (pH 6.0) was applied for HPLC

sy stem.

Calibration curves for voriconazole

Stock solutions of voriconazole (1 Clg/ Cll) were prepared in distilled water and

diluted in Lactated Ringer' s solution or plasma to give 100 Clg/ ml. Standards were

prepared by adding the diluted voriconazole solution to appropriate volumes of Lactated

Ringer' s solution or plasma to give a concentrations of 0.025, 0.1i, 0.2, 0.5, 1, 2, 4, 8 and

16 Clg/ ml. The plasma standards were also extracted before inj ecting to HPLC system.

Calibration curves were constructed by plotting the peak area of voriconazole against

concentration using a weighted (1/X2) least squares regression for HPLC data analysis.

Results

The HPLC data printout from the integrator was transferred into EXCEL 2003

(Microsoft Corporation) spreadsheets. With the equation obtained from the Weighted

(1:X2) Linear Regression, the voriconazole concentrations rom Lactated Ringer' s

solution, rat and human plasma samples were calculated [132]









Stability of Voriconazole in Lactate Ringer's Solution and Plasma

The in vitro stability of voriconazole in Lactate Ringer' s solution and human and

rat plasma were studied. The results showed that voriconazole in Lactate Ringer' s

solution, human and rat plasma were found stable at -800C for at least 6 months, at -40C

at least for 72 hours, room temperature and 370C for at least 24 hours which is consistent

with previous report 14' 30]

Protein Binding

The results of recovery for both extraction efficiency (EE) and retrodialysis (RD)

methods were summarized in Table 5-2 and 5-3. As shown from the tables, the in vitro

recovery of this compound from both extraction efficiency (EE) and retrodialysis

methods (RD) were quite similar with 51.1% + 2.6% (mean & SD) (ranging from 46.2%

to 56.3%) for EE method and 51.9% + 2.9% (mean & SD) (ranging from 45.3% to

57.8%) for RD method. The results of protein binding of voriconazole in presence of

different drug concentrations in rat and human plasma were summarized in Table 5-4 and

5-5. The protein binding of voriconazole in rat plasma is 66.2% + 2.2% (mean & SD)

(ranging from 61.5% to 70.3%) and that in human plasma is 57.0% + 2.8% (mean & SD)

(ranging from 50.9% to 62.1%).

Discussion

This experiment confirmed that the performance of voriconazole microdialysis was

feasible. It has been reported in the literature that recoveries were dependent on the

method used for the determination as well as on the flow rate, but independent of drug

concentration [133]. In Our current study, the recoveries of voriconazole were independent

of both the methods (EE and RD) and drug concentrations.









Only the free or unbound fraction of drug is available for antimicrobial activity

according to the "free drug hypothesis" of anti-infective agents. Voriconazole exhibits

moderate plasma protein binding with values of 66% in rat plasma and 58% in human

plasma, which was measured by 14C Voriconazole 131]. The binding is independent of

dose or plasma drug concentrations [s. Our current study is the first time to report the

voriconazole protein binding with in vitro microdialysis method as our knowledge. The

average plasma protein binding of voriconazole in rat and human is 66.2% and 57.0%,

respectively, which is consistent with the previous publicationS [8, 31]. Moreover, we also

confirmed that the protein binding in rat and human plasma was independent of the drug

concentrations. The free drug concentrations provided a better correlation to outcome

than total drug concentrations. Free drug concentrations must be considered when

examining the relationship between pharmacokinetic parameters and in vivo activity.

Microdialysis provided a simple, clean method for determining the free drug

concentration in vitro and in vivo.
























A


__ __ _


Table 5-1. List of materials
Materi al s
Vori conazol e
Acetonitrile
Ammonium Phosphate
Lactated Ringer's solution (USP)
Rat plasma
Human plasma
CMA/20 probe
Balance AB104
Vortex
Pipette tips
Tubes microcentrifuge
Centrifuge


Resource
Pfizer, New York, NY
Fisher, HPLC grade, Pittsburgh, PA
Sigma, HPLC grade, St. Louis, MO
Abbott, Chicago, IL
Wistar, Harlan Sprague-Dawley, Indianapolis, IN
Civitan Regional blood system, Gainesville, FL
Stockholm, Sweden
Mettler, Toledo, Hightstown, NJ
Kraft Apparatus model PV-5, Fisher, Pittsburgh, P
Fisher, Pittsburgh, PA
Fisher, Pittsburgh, PA
Fisher model 235V, Pittsburgh, PA










Table 5-2. Extraction efficiency (EE) method for measuring voriconazole recovery
Recovery Concentration (pp/ml) Average
(%) 0.51 1.04 2.06 5.13 10.31 41.47
1 50.6 50.7 46.8 56.3 52.6 50.6
2 46.6 51.8 53.4 52.8 54.4 46.1
3 46.2 52.0 52.2 51.5 56.2 49.6
Mean (%) 47.8 51.5 50.8 53.5 54.4 48.8 51.1
SD (%) 2.4 0.7 3.5 2.4 1.8 2.3 2.6













Table 5-3. Retrodialysis (RD) method for measuring voriconazole recovery
Recovery Concentration (pgml) Average
(%) 0.52 0.99 1.96 5.12 10.39 40.37
1 49.9 51.6 49.4 56.2 57.8 47.6
2 47.1 51.8 53.5 52.3 55.7 51.5
3 45.3 53.2 53.7 54.3 52.7 51.2
Mean (%) 47.4 52.2 52.2 54.3 55.4 50.1 51.9
SD (%) 2.3 0.9 2.5 1.9 2.6 2.2 2.9










Table 5-4. Rat plasma protein binding data of voriconazole measured by in vitro
microdialysis
Protein binding Concentration (pp/ml) Average
(%) 0.53 1.13 2.22 5.42 11.09 41.19
1 61.5 64.9 66.0 65.8 66.9 70.3
2 62.4 66.7 68.5 68.5 69.2 63.2
3 61.7 69.1 63.3 68.4 68.7 65.7
Mean (%) 61.9 66.9 65.9 67.5 68.2 66.4 66.2
SD (%) 0.4 2.1 2.6 1.5 1.2 3.6 2.2















Table 5-5. Human plasma protein binding data of voriconazole measured by in vitro
microdialysis
Protein binding Concentration (pg/ml) Average
(%) 0.54 1.09 2.16 5.55 11.15 41.13
1 50.9 51.8 56.5 57.6 61.2 62.1
2 55.9 55.9 56.6 60.2 60.8 60.3
3 56.8 54.4 51.5 59.8 55.9 57.9
Mean (%) 54.6 54.0 54.9 59.2 59.3 60.1 57.0
SD (%) 3.2 2.1 2.9 1.4 2.9 2.1 2.8









CHAPTER 6
IN VIVO PHARMACOKINETIC STUDIES OF VORICONAZOLE INT RATS

Background

Voriconazole is an antifungal agent which has been shown to be effective in the

treatment of patients with candidiasiS [134-136]. The effectiveness of voriconazole and other

antifungal agents is determined by the sensitivity of the fungi and the concentrations

achieved at the site of infection [34, 131, 137] However, the site of infection is generally not

the blood. Therefore, in many infections concentrations in organ tissues, extracellular

fluids, and inflammatory cells rather than plasma concentrations determine the clinical

outcome and may allow a better prediction of the therapeutic effect than plasma

concentrations [73, 74, 95, 99, 138-141]. Microdialysis has been used to study drug free levels in

many tissues and organs, such as muscle, adipose tissue, bile, blood, eye, skin, brain,

lung, and so on [73, 74]. It is an established technique for studying physiological,

pharmacological, and pathological changes of many drugs such as antibiotiCS [142, 143]

anti-cancer drugs [9~1416, and psychoactive compounds [17 4-4]in both preclimical

and clinical studies.

Specific Aims

The obj ective of this proj ect is to study the distribution profile of voriconazole in

the body by measuring its concentration in plasma, muscle. To evaluate unbound muscle

concentrations of voriconazole in Wistar rats by microdialysis after intravenous

administration of voriconazole and to compare the free tissue concentration to free

plasma concentration. To develop a population PK model of voriconazole in rats using

NONMEM@.












Reagents and Equipment

All of them have been described in Table 5-1 in Chapter 5 unless otherwise stated.

Isoflurane, USP was distributed by Webster Veterinary. Supply, Charlotte, NC

Animals

Wistar male rats (Harlan Sprague-Dawley, Indianapolis, IN) were used in the

proj ect.

Previous studies with other azole drugs have indicated that rat is a suitable test

model to investigate tissue distribution using microdialysiS [133, 137, 147, 148]. The rats were

shipped to ACS a couple of days before the study starts in order to reduce stress and

adapt them to the researcher. In the experiment, the rats were weighed before the surgery

and dose administration. The body weight of male Wistar rats used in this experiment

was ranged from 350g to 400g and age of 2-4 months. The animals were numbered in the

sequence of the experiments without identifying devices such as tattoos or collar numbers

since this project involved non-survival surgical experiment.

Experimental Design

Two groups of anesthetized male Wistar rats of six each were administered

intravenously with either 5mg/kg or 10 mg/kg dose of voriconazole. Total plasma

concentrations as well as unbound concentrations in thigh muscle were measured over 8-

hour period by microdialysis. Animals were sacrificed at the end of the study. The data

were analyzed by both non-compartmental and compartmental pharmacokinetic

approaches.

The study was performed with the approval of the Institutional Animal Care and

Use Committee (IACUC) of the University of Florida (Protocol # F109). The animals


Materials and Methods









were housed in the Animal Care Services (ACS) at the Health Science Center, University

of Florida according to the standard husbandry procedures.

Anesthetic Procedure

Isoflurane has been used in the surgical procedure for anesthesia in rats in many

reports [150-153]. In Our studies, the rats were anesthetized by using Isotec-4 isoflurane

vaporizer (SurgiVet/ Smiths Medical, Waukesha, WI) from ACS. After placing the rat in

the induction chamber, the flowmeter was adjusted to 0.8-1.5L/min and the isoflurane

vaporizer was adjusted to 2% which was marked as "2". Anesthesia was confirmed by

the absence of reflexes after pinching the rat' s footpads. After anesthesia, the rat was

immobilized in a supine position on a dissecting board by holding the paws with rubber

bands, which was attached to pins on each corner of the board. For maintenance of

anesthesia, a mask connected to the Bain Circuit was used for the rat, and the flowmeter

was adjusted to 400-800mL/min. The footpads were pinched in a regular basis (every 15

min) during the experiment. The rat was kept normothermic on an electric heating pad.

Antiseptic Procedure

This project involved non-survival surgical experiment. The surgery sites were

disinfected by swiping the area with 70% isopropyl alcohol and skin areas of surgery

sites were shaved before surgery.

Blood Samples

Blood sample were collected by a polyethylene catheter (inner diameter of 0.3 mm

and outer diameter of 0.7 mm) introduced in the carotid. After initial insertion and

following each sampling, the catheter was irrigated with 300C1l 100 IU pre-warmed

heparinized saline. Blood samples (100C1l) were collected in heparinized tubes before the

dose was administered (time zero) and at 10, 30, 60, 90, 120, 180, 240, 300, 360, 480min









after drug administration. The blood samples were centrifuged at approximately 2000

rpm for 10 minutes at room temperature and the supernatant of plasma was transferred to

a new Eppendorf tube. This step was repeated one more time. Then plasma samples were

kept at -80 oC until analyzed.

Microdialysis

Probe calibration

The left hind leg muscle was used for insertion of a microdialysis probe after skin

area was shaved and cut open. Before introducing the probe itself, a guide plastic cannula

was introduced in the muscle with the help of a 23-guage needle. Afterwards, the plastic

cannula was kept in place and the needle was removed and replaced by the microdialysis

probe. After inserting the probe, the plastic cannula was removed by tearing it out while

holding the probe in place.

Probe calibration was performed before drug administration. The probe was

initially perfused with lactated Ringer' s solution (NaCl 137 mM, KCl 1.0 mM, CaCl2 0.9

mM, NaHCO3 1.2 mM) at a flow rate of 2 Cll /min for 30 minutes using a Harvard

Apparatus 22 inj section pump, model 55-4150. After equilibration, the probe was

calibrated by retrodialysis. In this method, the syringe with lactated Ringer' s solution was

replaced by a syringe with a voriconazole solution of 1 or 2Cpg/ml. The drug solution was

initially pumped at a flow of 5 Cll/min for 5 min and after the flow was changed back to 2

Cll/min. The probe was allowed to equilibrate for 0.5 hour before samples were collected.

The samples for the calibration procedure were collected very 20 minutes and a total of 3

samples were collected in each experiment. The samples were frozen at -80 oC before

analyses by a validated HPLC/UV method.









Muscle microdialysis

After the probe calibration, the drug perfusion was stopped and removed instead of

blank lactated Ringer's solution. The probe was perfused with lactated Ringer's solution

for 30 minutes to wash out. After wash out, microdialysis samples were collected for

muscle over 20-minute intervals at times 20, 40, 60, 80, and up to 480 minutes after drug

admini strati on.

The in vivo recovery was calculated by the same above equation of

C -C
R% = ,,ft rx 100 (6-1)
perfusate

The drug concentration in the tissue was calculated using the equation of


C dialysate
it, pee =(6-2)
R%

Cr,fe is the muscle free concentration, Cdralysate is the concentration in the dialysate,

and R%/ is the percent of recovery obtained in each animal experiment.

Data Analysis

Microdialysis samples were measured using HPLC/UV method directly. Plasma

samples were extracted before the measurement using HPLC/UV method as being

described previously (Methods in Chapter 5) with modification. Briefly, plasma samples

were thawed on ice and then centrifuged at 1,700 x g for 5 min. An aliquot (25 CIl) was

pipetted into a 0.7 ml Eppendorf tube and acetonitrile (40 Cll) was then added. The

mixture was mixed briefly by vortex, centrifuged at 1,200 x g for 5 min after standing for

10 min at room temperature and supernatant was transferred to a new Eppendorf tube.

This step was repeated one more time. Then 25 Cll of the supernatant plus 25 Cll of 0.04 M

Ammonium Phosphate buffer (pH 6.0) was applied for HPLC system.









The HPLC data was input into EXCEL 2003 (Microsoft Corporation) spreadsheets.

With the equation obtained from the Weighted (1:X2) Linear Regression, the rat and

human plasma sample concentrations were calculated [132]. With the equation obtained

from the Weighted (1:X2) Linear Regression function, the rat plasma sample

concentrations were calculated. Unbound concentrations in rat thigh muscle were

measured by microdialysis. Individual probe recovery was determined by retrodialysis

and allowed conversion of the measured dialysate concentrations to the actual unbound

tissue concentrations. The data were analyzed both by noncompartmental and

compartmental pharmacokinetic approaches.

Noncompartmental pharmacokinetic analysis

The following parameters were determined for each rat for both plasma and

microdialysis noncompartmental pharmacokinetic analysis. The mean and standard

deviation (SD) of each parameter were also determined.

The data were analyzed by WinNonlin 5.2 (Pharsight Corporation, Mountain

View, CA) to describe the best absorption model. The initial concentration Co was

determined by logarithmic back-extrapolation to t = 0 using the first two data points. The

terminal elimination rate constant (ke) was determined by linear regression of the log

plasma concentrations. The terminal half-life (tl/2) WAS calculated as In(2)/ke. The area

under the concentration-time curve (AUCt) values were calculated between time 0 and

the final time point at which measurable drug concentrations were observed, using the

linear trapezoidal rule. AUCoo was calculated by extrapolation from time zero to infinity

with ke. The area under the first moment curve (AUMC) was calculated from a plot of Ct

versus t using the trapezoidal rule up to the last data point (Cx) at time tx and adding the

extrapolated terminal area, calculated as Cx-tx/ke + Cx/ke2. AUMCoo was Area under the









first moment curve when the time concentration curve is extrapolated to infinity. The

mean residence time (MRT) was calculated as AUMC/AUC. Clearance (CL) was

calculated by the relationship of Dose/AUCoo, and the volume of distribution was

calculated by the relationship CL/ke. The volume of distribution at steady state (Vdss)

was calculated as (Dose-AUMC)/AUC2.

Compartmental pharmacokinetic analysis and modeling

Both the individual and average plasma and free tissue concentrations at each

time point were fitted by using the modeling software of NONMEM" (UCSF, San

Francisco, CA).

Rat total plasma voriconazole pharmacokinetic data were first analyzed using

one- and two-compartment model with linear or nonlinear elimination (Michaelis-Menten

elimination). Then rat total plasma and unbound muscle voriconazole data were analyzed

simultaneously using a two-compartment model with nonlinear elimination. The

schemes of one- and two-compartment model with nonlinear elimination were shown in

Figure 6-1 and Figure 6-2. The models were fitted to the data using the first order (FO)

method and the subroutine ADVAN1 TRANS2 (one-compartment with linear

elimination), ADVAN3 TRANS4 (two-compartment with linear elimination),

ADVAN10 TRANS 1 (one-compartment with nonlinear elimination), and ADVAN6

TRANS 1 (two-compartment with nonlinear elimination). NONMEM performs linearized

maximum likelihood estimation by use of an obj ective function (OF). Typical values

(population means) with their corresponding standard errors (SE) and the intersubject and

intrasubj ect variabilities were expressed as coefficients of variation (CV percentage). To

determine whether there was a statistically significant difference between the goodness of

fit between the two models, the Akaike model selection criteria was used, which required









a decrease of two points in the obj ective function (minus twice the logarithm of the

likelihood of the model) to accept a model with one additional parameter, as well as a

comparison of diagnostic plots (observed concentrations compared with predictions and

observations/predictions compared with time).

The one-compartment model with linear elimination will be parameterized in

terms of CL (clearance), and V (volume of central compartment). The two-compartment

model with linear elimination will be parameterized in terms of CL (clearance), V1

(volume of central compartment), V2 (VOlume of peripheral compartment), and Q

(intercompartmental clearance).

The one-compartment model with nonlinear elimination will be parameterized in

terms of Vmax (maximum elimination rate), Km (the Michaelis-Menten constant,

voriconazole concentration at which the elimination is at half maximum) and V (volume

of central compartment). The two-compartment model with nonlinear elimination will be

parameterized in terms of Vmax (maximum elimination rate), Km (the Michaelis-Menten

constant, voriconazole concentration at which the elimination is at half maximum), V1

(volume of central compartment), V2 (VOlume of peripheral compartment), and Q

(intercompartmental clearance).

The differential equations for a two-compartment model with nonlinear

elimination (Michaelis-Menten elimination) are given in following equations:

A (1)
Cl = (6-3)
Vl



= 2 1 A 2 ) V ~ a x C 11 l 2 A ( 1 ) ( 6 4 )









dA (2)
= kl2 A(1) k21 -A(2) (6-5)



Kl2 = 7~(6-6)


K21 = (6-7)
V2

Where A(1) represents the amount of voriconazole in the central compartment,

A(2) represents the amount of voriconazole in the peripheral compartment, Vmax

represents the maximal elimination rate, Km is the Michaelis-Menten constant

(voriconazole concentration at which the elimination is at half maximum), C1 is the

predicted serum concentration of voriconazole, and kl2 and k21 are the

intercompartmental rate constants. V1 is the volume of central compartment, V2 is the

volume of peripheral compartment, and Q is the intercompartmental clearance.

Because free rather than protein-bound voriconazole was measured in dialysate, a

scaling factor for the tissue compartments was introduced, accounting for the free

fraction of voriconazole, fu.

A(2)
C2 = x .fu (6-8)
V2

Interindividual variability for the pharmacokinetic parameters was modeled_using

the following exponential error model:

Vmax,= Ox-EXP(r,) (6-9)

K, = 82 EXP(T2) (6-10)

Y~= 83 EXP(r3) (6-11)

Q = 84 EXP(T4) (6-12)









V2 85' EX 5)


(6-13)


Where 6 is the population mean estimate (or typical value) of the corresponding

pharmacokinetic parameter, and rli's are the associated interindividual variability. The rl

values are independent, identically distributed random errors with mean of zero and a

variance equal to coi2. Interindividual variability will be described by an exponential error

model .

When analysising the total plasma only, we used a proportional error model for the

intraindividual residual variability of plasma concentration for estimation, which was

described by the following equation:

Cp = p -(+Sy~rop)(6-14)

where Cpij is the observed value of the jth' plasma concentration of individual i; Opij is the

predicted jth plasma concentration of individual i; and Si,; prop denote the proportional

residual random errors distributed with zero means and variance G2 prop.

When analysising the total plasma and unbound muscle voriconazole data

simultaneously, the intraindividual residual variability of total plasma concentration were

estimated using a proportional model, as described previously and the unbound muscle

data were using a combination of proportional and additive error model, as described by

the following equation:

CP3, = CP3, (1+ I,,,pro)+ U,add (6-15)

where Cpij is the observed value of the jth plasma concentration of individual i; Opij is the

predicted jth plasma concentration of individual i; and Si, prop and Siij,add denote the

proportional and additive residual random errors distributed with zero means and

vainc 2 an 2 aa









Plasma total voriconazole concentrations and unbound muscle voriconazole

concentrations were assigned separate residual errors, e1, E2 and 83, TOSpectively. The

following criteria were considered to determine the best model:

In the case of nested models (one model is a subset of other, i.e., null model

(without covariates) is a subset of full model (with covariates), the minimum obj ective

function value (OFV) of the best model should be significantly smaller than the

alternative models) based on the maximum likelihood ratio (MLR) test. The MLR test

was applied when the test models fulfilled the full/reduced model definition. A full

model can be made equivalent to a reduced model by setting a parameter to a fixed value.

The change in OFV between the two nested models is approximately X2 distributed with

degree of freedom equal to the number of parameters that are set to a fixed value in the

reduced model. A decrease of 3.84 units in the OFV will be considered statistically

significant (p < 0.05) for addition of one parameter during the development of the model.

In the case of non-nested models, the different structure models were evaluated

based on minimization of the Akaike information criterion (AIC) value. The AIC was

defined in terms of:

AIC = OFV + 2*p (6-16)

where p = total number of parameters in the model (structural + error). Lower the value

of AIC better the model.

A plot of voriconazole concentration (observed (DV), individual predicted

(IPRED) and population predicted (PRED)) versus time for all the individual were

drawn, which would give an overall trend of fitted concentrations. It could be seen that

for certain individuals the population predictions are underpredicted or overpredicted.









The observed and predicted plasma concentrations would be more randomly distributed

across the line of unity for the preferred model. NONMEM obtains IPRED values by the

Bayesian POSTHOC option. Using the population mean estimate of parameters (prior)

and each individual data (likelihood), NONMEM obtains the individual parameter

estimates (posterior). From the individual parameter estimates, individual predicted

concentrations (IPRED) are obtained.

A graph of DV vs. IPRED and DV vs. PRED (goodness of fit plots) could be

looked for any bias in the predictions. Ideally the points should be uniformly distributed

along the line of identity. Normally DV vs. IPRED is much better than DV vs. PRED as

PRED contains unexplained variability.

Plots of the residuals (RES) vs. PRED and weighted residuals (WRES) vs. PRED

(goodness of fit plots) could be looked for any unaccounted heterogeneity in the data.

The NONMEM codes were written and the initial estimation of all parameters

obtained by WinNonlin was used to run NONMEM. The figures were generated with S-

PLUS (Statistical Sciences, Version 6.2).

Results

Probe Recovery

The individual probe recovery in each experiment was used to convert the

microdialysate concentrations to unbound tissue concentrations and all the values were

summarized in Table 6-1. As shown from the tables, the in vivo recovery of voriconazole

from retrodialysis methods (RD) were quite similar with 43.3 & 4.2% (mean & SD) (range

from 37.6% to 48.1%) for 1 Clg/ml group and 47.6 & 4.0% (mean & SD) (range from

43.4% to 53.6%) for 2 Clg/ml group.









Individual Pharmacokinetic Analysis of Total Voriconazole in Plasma after
Intravenous Bolus Administration

The experimental data of total plasma voriconazole from 5 and 10 mg/kg

intravenous dose were listed in Table 6-2 and 6-3 and were analyzed by both

noncompartmental pharmacokinetic analysis and compartmental pharmacokinetic

analy si s.

Noncompartmental pharmacokinetic analysis

After noncompartmental pharmacokinetic analysis by WinNonlin 5.2, the results of

voriconazole for the total plasma data after intravenous bolus administration of 5 and 10

mg/kg were listed in Table 6-2 and 6-3. The initial concentrations (CO) were 3.78 & 0.29

mg/L (mean & SD) and 7.76 & 0.18 mg/L (mean & SD) respectively, which declined with

a terminal half-life of 3.95 & 0.34 hours (mean & SD) and 4.92 & 0.29 hours (mean & SD).

The areas under the curve AUCoo were 20.53 & 2.57 hr-mg/L (mean & SD) and 56.07 &

3.82 hr-mg/L (mean & SD), respectively. The areas under the first moment curve

(AUMCoo) were 121.90 & 24.64 hr2-mg/L (mean & SD) and 417.35 & 43.75 hr2-mg/L

(mean & SD), respectively. The residence times (MRToo) were 5.89 & 0.52 hours (mean &

SD) and 7.43 & 0.40 hours (mean & SD), respectively. The volumes of distribution of the

central compartment (Vc) were 1.33 & 0.1 mL/g (mean & SD) and 1.29 & 0.03 mL/g

(mean & SD), respectively and Vdss were 1.44 & 0.09 mL/g (mean & SD) and 1.33 & 0. 12

mL/g (mean & SD), respectively. The total body clearances (CL) were 0.25 & 0.03

mL/hr/g (mean & SD) and 0.18 & 0.02 mL/hr/g (mean & SD), respectively.

Compartmental pharmacokinetic analysis

The experimental data of total plasma voriconazole from 5 and 10 mg/kg

intravenous dose were plotted in Figure 6-3 and 6-4 and were analyzed by both one- and










two-compartment model with linear or nonlinear elimination (Michaelis-Menten

elimination) by NONMEM. Initially, all four models have applied to fit unbound plasma

and muscle voriconazole PK data from 5 and 10 mg/kg intravenous dose, and the OFV

and AIC value have been listed in Table 6-4. As shown in this table, the AIC value (-

344.03) from two-compartment model with nonlinear elimination was the lowest, which

indicated that this model might be the best to fit the total plasma voriconazole PK data in

rats.

One-compartment model with linear elimination analysis

A one-compartment model with linear elimination was used to fit the data of total

plasma voriconazole from 5 and 10 mg/kg intravenous dose and was able to produce a

good curve fit of these data. As shown in Table 6-4, the OFV is -240.73, and the AIC is -

230.73, which is the highest compared to other models. The critical parameters obtained

in this analysis were listed in Table 6-5. The diagnostic plots for this one-compartment

model were also shown in Figure 6-5, 6-6 and 6-7. As shown in Figure 6-5, the observed

and predicted plasma concentrations were randomly distributed across the line of unity.

As shown in Figure 6-6, the points of DV, IPRED and PRED were uniformly distributed

along the line of identity, and in Figure 6-7, the residuals and weighted residuals plot

versus predicted voriconazole concentration for this model showed a relatively uniform

distribution of residuals over the concentration range.

Two-compartment model with linear elimination analysis

A two-compartment model with linear elimination was also used to fit the data of

total plasma voriconazole from 5 and 10 mg/kg intravenous dose and was able to produce

a good curve fit of these data. As can be seen from Table 6-4, the OFV is -295.77, and

the AIC is -277.77. The critical parameters obtained in this analysis were listed in Table









6-6. The diagnostic plots for this two-compartment model were also shown in Figure 6-8,

6-9 and 6-10. The observed and predicted plasma concentrations were randomly

distributed across the line of unity except some early time point data. The points of DV,

IPRED and PRED were uniformly distributed along the line of identity, and the residuals

and weighted residuals plot versus predicted voriconazole concentration for this model

showed a relatively uniform distribution of residuals over the concentration range.

One-compartment model with nonlinear elimination

A one-compartment model with nonlinear elimination was used to fit the data of

total plasma voriconazole from 5 and 10 mg/kg intravenous dose and was able to produce

a good curve fit of these data. As shown in Table 6-4, the OFV is -249. 17, and the AIC is

-23 5.17. The critical parameters obtained in this analysis were listed in Table 6-7. The

diagnostic plots for this one-compartment model were also shown in Figure 6-11, 6-12

and 6-13. As shown in Figure 6-11i, the plot of voriconazole concentration (observed

(DV), individual predicted (IPRED) and population predicted (PRED)) versus time for

the entire individual gave an overall trend of fitted concentrations. The observed and

predicted plasma concentrations were randomly distributed across the line of unity. As

shown in Figure 6-12, the points of DV, IPRED and PRED were uniformly distributed

along the line of identity, and in Figure 6-13, the residuals and weighted residuals plot

versus predicted voriconazole concentration for this model showed a relatively uniform

distribution of residuals over the concentration range.

Two-compartment model with nonlinear elimination

To find the best model to fit the data of total plasma voriconazole from 5 and 10

mg/kg intravenous dose, a two-compartment model with nonlinear elimination was also

applied. This model was able to produce a good curve fit of these data. As shown in









Table 6-4, the OFV is -366.03, and the AIC is -344.03, which is the lowest compared to

other models. The critical parameters obtained in this analysis were listed in Table 6-8.

The diagnostic plots for this two-compartment model were also shown in Figure 6-14,

6-15 and 6-16. As shown in Figure 6-14, the plot of voriconazole concentration (observed

(DV), individual predicted (IPRED) and population predicted (PRED)) versus time for

the entire individual gave an overall trend of fitted concentrations. The observed and

predicted plasma concentrations were randomly distributed across the line of unity. As

shown in Figure 6-15, the points of DV, IPRED and PRED were uniformly distributed

along the line of identity. As can be seen in Figure 6-16, the residuals and weighted

residuals plot versus predicted voriconazole concentration for this model showed a

relatively uniform distribution of residuals over the concentration range. All the

diagnostic plots and AIC value have indicated that the two-compartment model with

nonlinear elimination seem to adequately explain the variability of the data and fit the

data better than other models.

Individual Pharmacokinetic Analysis of Unbound Voriconazole in Muscle after
Intravenous Bolus Administration

The experimental data of unbound muscle voriconazole concentration from 5 and

10 mg/kg intravenous dose were obtained from rats muscle in vivo microdialysis based

on in vivo recovery of voriconazole from retrodialysis methods (RD) as described

previously in this chapter. All the data were listed in Table 6-9, 6-10 and were analyzed

by both noncompartmental pharmacokinetic analysis and compartmental

pharmacokinetic analysis.









Noncompartmental pharmacokinetic analysis

After noncompartmental pharmacokinetic analysis, the results of voriconazole for

the unbound muscle data after intravenous bolus administration of 5 and 10 mg/kg were

listed in Table 6-9 and 6-10. The values of terminal half-life were 2.83 & 0.20 hours

(mean & SD) and 6.92 & 2.29 hours (mean & SD). The areas under the curve AUC, were

7.52 & 1.02 hr-mg/L (mean & SD) and 24.52 & 7.84 hr-mg/L (mean & SD), respectively.

The areas under the first moment curve (AUMC,) were 36.72 & 6.98 hr2-mg/L (mean &

SD) and 267.98 & 184.84 hr2-mg/L (mean & SD), respectively. The residence times

(MRT,) were 4.85 & 0.31 hours (mean & SD) and 10. 16 & 3.02 hours (mean & SD),

respectively. The total body clearances (CL) were 0.68 & 0.10 mL/hr/g (mean & SD) and

0.43 & 0. 11 mL/hr/g (mean & SD), respectively.

Compartmental pharmacokinetic analysis

Since the two-compartment model with nonlinear elimination is the best model to

fit the total plasma voriconazole PK data in rats, the experimental data of total plasma

and unbound muscle voriconazole PK data from 5 and 10 mg/kg intravenous dose were

analyzed simultaneously by two-compartment model with nonlinear elimination

(Michaelis-Menten elimination) by NONMEM. The experimental original unbound

muscle voriconazole PK data were plotted in Figure 6-17 and 6-18.

Two-compartment model with nonlinear elimination

To find the best model to fit the data of unbound plasma and muscle voriconazole

from 5 and 10 mg/kg intravenous dose, a two-compartment model with nonlinear

elimination was applied. This model was also able to produce a good curve fit of these

data. The critical parameters obtained in this analysis were listed in Table 6-11. The

diagnostic plots for this model were also shown in Figure 6-19, 6-20 and 6-21. As shown









in Figure 6-19, the plot of total plasma and unbound muscle voriconazole concentration

(observed (DV), individual predicted (IPRED) and population predicted (PRED)) versus

time for the entire individual gave an overall trend of fitted concentrations. The observed

and predicted plasma concentrations were randomly distributed across the line of unity.

As shown in Figure 6-20, the points of DV, IPRED and PRED were uniformly distributed

along the line of identity. As can be seen in Figure 6-21, the residuals and weighted

residuals plot versus predicted voriconazole concentration for this model showed a

relatively uniform distribution of residuals over the concentration range. It indicated that

the two-compartment model with nonlinear elimination seem to adequately explain the

variability of the unbound plasma and muscle voriconazole PK data.

Pharmacokinetic Analysis of Average Total Voriconazole Data in Plasma and
Unbound Voriconazole Data in Muscle after Intravenous Bolus Administration

The average data of total plasma and unbound muscle voriconazole from 5 and 10

mg/kg intravenous dose were listed in Table 6-2, 6-3, 6-9 and 6-10, and were fitted

simultaneously by two-compartment model with nonlinear elimination (Michaelis-

Menten elimination) by NONMEM using the respective parameters of Vl= 0.5 L, Q =

0.28 L/hr, Vmax = 0.34 mg/hr, Km = 1.52 mg/L. This model produced a good curve fit of

the average plasma and unbound muscle voriconazole concentrations (Figure 6-10 and 6-

11).

The data of unbound plasma voriconazole concentration from 5 and 10 mg/kg

intravenous dose were obtained from total plasma concentration based on a protein

binding of 66.2% in rat plasma, which was measured by in vitro microdialysis as

described in Chapter 5. The average data of unbound plasma voriconazole were also

calculated and plotted in the same figures (Figure 6-10 and 6-11).









Discussion

In this study, the pharmacokinetics of voriconazole in male Wistar rats in doses of

5 and 10 mg/kg given intravenously was investigated. Plasma concentrations as well as

unbound tissue concentrations in thigh muscle were measured. As it is known, antifungal

effect relates to the antifungal agent concentrations in plasma and tissue at the target site.

The aim was to establish the relationship between plasma concentration and tissue

concentration, which may allow us to predict the pharmacodynamic effect of antifungal

agent and help optimize the dosage regimen.

Elimination of voriconazole from plasma does not follow simple linear

pharmacokinetics in rats which is consistent with the previous report 131]. The

voriconazole plasma profiles following intravenous administration (Figure 6-3, 6-4) are

convex to some extent indicating the characteristic of compounds that show capacity-

limited elimination. Compared the 5 mg/kg dosing regimen to 10 mg/kg,

pharmacokinetic parameters are dependent upon dose. There was superproportional

increase in area under the curve was seen with increasing dose in rat studies. There is 2.7-

fold increase in AUC for a 2-fold increase in intravenous dose. It has been reported that

voriconazole was eliminated predominantly by metabolism such as the human hepatic

cytochrome P450 enzymes of CYP2C 19, CYP2C9 and CYP3A4 131]. The terminal

elimination half-lives calculated for voriconazole in was 3.95h for 5mg/kg dosage and

4.92h for 10mg/kg dosage, which is similar to what has been reported in the literatures

for rats [31, 154]. In Arauj o, et al's report, the estimated half-life was found to be 2.4+0.6 h

considering that the pharmacokinetic study was carried out for 8 h. In Roffey, et al's

report, the voriconazole's plasma PK data were extracted and concentrations were fitted










to a noncompartment analysis (NCA) model using WinNonlin. The estimated half-life

was found to be 4 h considering that the pharmacokinetic study was carried out for 6h.

The elimination of voriconazole was characterized by nonlinear pharmacokinetics

in rats. It is likely that saturation of metabolic clearance is the cause of the nonlinearity.

Compared the fit of linear with nonlinear models, 1-compartment with 2-compartment

models, the 2-compartment with nonlinear elimination had the best fitting with a lowest

AIC value of -3 44.03.

Microdialysis has been reported to be a very useful technique to measure a variety

of drugs in different tissues or endogenous compounds in vivO [73, 133, 155]. It is very

important to determine the probe recovery when applying microdialysis for a new

compound if a good prediction of the true tissue levels is wanted. The recoveries of

voriconazole in rat muscle were determined by retrodialysis method in our studies.

Voriconazole exhibits moderate plasma protein binding and is independent of

plasma drug concentrations in vitro. The unbound concentrations of voriconazole in the

muscle were measured directly and compared with the calculated free plasma

concentrations using the average plasma protein binding of voriconazole in rat of 66.2%.

The results showed that muscle had similar concentrations of unbound plasma

concentrations.

We fitted plasma and tissue concentrations simultaneously. A two-compartment

model with nonlinear elimination described the data well and resulted in a value of fu

(0.38) for muscle. The fu can be interpreted as the free fraction of voriconazole in tissue

such as in muscle which might be affected by the subj ect' s covariates, weight, height, and

the plasma albumin concentration [156, 157]










In summary, a two-compartment model with non-linear elimination was able to

produce a good curve of both the individual and the average plasma and muscle

voriconazole concentrations. The unbound voriconazole concentrations in muscle were

almost identical with the calculated unbound plasma concentrations for different dosages

indicating the muscle microdialysis sample may be used for estimating the free plasma

levels for voriconazole. The free fraction of voriconazole in tissue (fu) can also be

predicted by PK analysis.










Table 6-1. Recovery data of voriconazole using retrodialysis (RD) method in rats muscle
Animal Recovery (%)
1 Clg/ml 2 Clg/ml
1 39.9 53.6
2 41.5 46.9
3 37.6 51.0
4 46.3 43.4
5 48.1 43.8
6 46.1 47.1
Mean (%) 43.3 47.6
SD (%) 4.2 4.0





Table 6-2. Total plasma voriconazole (5 mg/kg) individual noncompartmental pharmacokinetics analysis


Time (hr)


Concentration (mg/L)


51
3.46
2.98
2.63
2.39
2.27
2.14
1.86
1.68
1.42
1.03


52 53
3.94 3.58
3.35 3.20
3.05 2.82
2.68 2.64
2.54 2.40
2.36 2.22
2.07 2.01
1.73 1.78
1.52 1.52
1.07 1.06
0.16 0.17
4.37 3.99
4.27 3.79
23.76 22.52
150.96 137.42
6.35 6.10
1.17 1.32
1.34 1.35
0.21 0.22


54
3.52
3.12
2.72
2.59
2.35
2.05
1.83
1.48
1.23
0.84
0.19
3.67
3.74
19.40
105.33
5.43
1.34
1.40
0.26


55 56 Mean SD Median
3.14 3.50 3.52 0.26 3.51
2.75 3.01 3.07 0.21 3.06
2.48 2.54 2.71 0.21 2.67
2.25 2.33 2.48 0.18 2.49
2.09 2.11 2.29 0.17 2.31
1.89 1.79 2.08 0.21 2.10
1.74 1.59 1.85 0.18 1.85
1.46 1.30 1.57 0.19 1.58
1.19 1.01 1.31 0.21 1.32
0.86 0.72 0.93 0.14 0.95
0.18 0.20 0.18 0.02 0.18
3.92 3.48 3.95 0.34 3.95
3.36 3.78 3.78 0.29 3.76
18.74 16.99 20.53 2.57 20.60
110.59 87.02 121.90 24.64 124.01
5.90 5.12 5.89 0.52 6.00
1.49 1.32 1.33 0.10 1.33
1.58 1.51 1.44 0.09 1.44
0.27 0.29 0.25 0.03 0.24


0.17
0.50
1.00
1.50
2.00
3.00
4.00
5.00
6.00
8.00


SKe [hr ] 0.16
tl/2 [hr] 4.26
Co [mg/L] 3.73
AUC,[(mg/L)-hr] 21.80
AUMC, [(mg/L)-hr2] 140.10
MRT, [hr] 6.43
V, [mL/g] 1.34
Vdss [mL/g] 1.47
CL [mL/hr/g] 0.23











Table 6-3. Total plasma voriconazole (10 mg/kg) individual noncompartmental pharmacokinetics analysis
Time (hr) Concentration (mg/L)
101 102 103 104 105 106 Mean SD Median
0.17 7.16 7.53 7.11 7.61 7.26 7.45 7.35 0.21 7.35
0.50 6.31 6.80 6.46 7.05 6.33 6.67 6.60 0.29 6.57
1.00 5.75 6.19 5.93 6.60 5.53 6.22 6.04 0.38 6.06
1.50 5.34 6.02 5.35 5.62 5.26 5.99 5.60 0.34 5.48
2.00 5.02 5.37 5.05 5.50 5.10 5.25 5.21 0.19 5.18
3.00 4.51 5.13 4.76 5.26 4.78 5.00 4.91 0.28 4.89
4.00 4.30 5.07 4.46 4.87 4.10 4.54 4.56 0.36 4.50
5.00 3.78 4.56 4.08 4.33 3.94 4.20 4.15 0.28 4.14
6.00 3.21 4.03 3.61 3.74 3.24 3.94 3.63 0.35 3.68
8.00 2.39 2.96 2.78 2.81 2.56 2.82 2.72 0.21 2.80
SKe [hr ] 0.15 0.15 0.13 0.14 0.14 0.14 0.14 0.01 0.14
tl/2 [hr] 4.56 4.76 5.40 4.81 4.95 5.03 4.92 0.29 4.88
Co [mg/L] 7.63 7.93 7.46 7.91 7.78 7.87 7.76 0.18 7.83
AUC,[(mg/L)-hr] 49.88 59.62 57.69 58.13 52.84 58.29 56.07 3.82 57.91
AUMC, [(mg/L)-hr2] 342.79 438.36 465.78 422.35 392.91 441.91 417.35 43.75 430.35
MRT, [hr] 6.87 7.35 8.07 7.27 7.44 7.58 7.43 0.40 7.39
V, [mL/g] 1.31 1.26 1.34 1.26 1.29 1.27 1.29 0.03 1.28
Vdss [mL/g] 1.38 1.22 1.50 1.18 1.37 1.32 1.33 0.12 1.34
CL [mL/hr/g] 0.20 0.17 0.19 0.16 0.18 0.17 0.18 0.01 0.18












Table 6-4. Comparison of obj ective function value (OFV) and Akaike information criterion (AIC) value from rats total plasma
voriconazole PK analysis using different models
Models Total number of parameters (p) Objective function value (OFV) Akaike information criterion (AIC) value
1-compartment with linear elimination 5 -240.73 -230.73

1-compartment with nonlinear elimination 7 -249. 17 -235.17

2-compartment with linear elimination 9 -295.77 -277.77

2-compartment with nonlinear elimination 11 -366.03 -344.03









Table 6-5. One-compartment model with linear elimination for rat total plasma
voriconazole PK analysis
Parameter estimates Population estimate (SE%) Between subj ect variability (BSV) (SE%)
CL (L/hr) 0.082 (8.7) 22.4% (37.3)
V (L) 0.58 (1.7) 5.1% (31.5)
Residual variability
Proportional error 5.5% (6.8)
Note: % SE: percent standard error of the population parameter estimate.













Table 6-6. Two-compartment model with linear elimination for rat total plasma
voriconazole PK analysis
Parameter estimates Population estimate (SE%) Between subject variability (BSV) (SE%)
CL (L/hr) 0.070 (6.7) 23.5% (28.3)
V, (L) 0.49 (1.7) 6.3% (28.0)
Q (Uhr) 0.35 (11.8) 10.0% (-)
V2 (L) 0.11 (7.9) 10.0% (-)
Residual variability
Proportional error 3.9% (12.1)
Note: Dashes indicate that the respective parameters tested during model building were
fixed. % SE: percent standard error of the population parameter estimate.









Table 6-7. One-compartment with non-linear elimination model for rat total plasma
voriconazole PK analysis
Parameter estimates Population estimate (SE%) Between subj ect variability (BSV) (SE%)
Vmax (mg/hr) 0.86 (24.8) 9.3% (74.7)
Km (mg/L) 4.55 (31.0) 1.0% (-)
V (L) 0.59 (1.5) 5.7% (27.5)
Residual variability
Proportional error 5.5% (5.9)
Note: Dashes indicate that the respective parameters tested during model building were
fixed. % SE: percent standard error of the population parameter estimate.











Table 6-8. Two-compartment with non-linear elimination model for rat total plasma
voriconazole PK analysis
Parameter estimates Population estimate (SE%) Between subject variability (BSV) (SE%)
V, (L) 0.50 (1.6) 6.5% (29.8)
Q (L/hr) 0.28 (8.6) 10.0% (-)
V, (L) 0. 14 (6.5) 11.4% (96.2)
Vmax (mg/hr) 0.34 (8. 1) 12.0% (42.4)
Km (mg/L) 1.52 (17.5) 10.0% (-)
Residual variability (SE%)
Proportional error 2.9% (15.9)
Note: Dashes indicate that the respective parameters tested during model building were
fixed. % SE: percent standard error of the population parameter estimate.











Table 6-9. Unbound muscle voriconazole (5 mg/kg) individual noncompartmental
pharmacokinetics analysis
Time (hr) Concentration (mg/L)
51 52 53 54 55 56 Mean SD Median
0.33 1.39 1.40 1.70 1.43 1.54 1.48 1.49 0.12 1.45
0.67 1.16 1.24 1.50 1.17 1.28 1.17 1.25 0.13 1.21
1.00 1.12 1.19 1.34 0.90 1.27 1.10 1.15 0.15 1.15
1.33 1.02 1.14 1.13 0.86 1.17 0.89 1.03 0.13 1.07
1.67 0.90 1.06 1.07 0.80 1.07 0.82 0.95 0.13 0.98
2.00 0.83 0.94 1.05 0.79 0.98 0.80 0.90 0.11 0.88
2.33 0.77 0.86 1.00 0.71 0.92 0.71 0.83 0.12 0.82
2.67 0.74 0.84 0.93 0.67 0.90 0.69 0.80 0.11 0.79
3.00 0.71 0.80 0.86 0.67 0.88 0.67 0.77 0.09 0.75
3.33 0.68 0.78 0.82 0.63 0.87 0.65 0.74 0.10 0.73
3.67 0.65 0.76 0.77 0.58 0.87 0.63 0.71 0.11 0.71
4.00 0.63 0.74 0.76 0.56 0.83 0.61 0.69 0.10 0.68
4.33 0.61 0.71 0.70 0.55 0.74 0.58 0.65 0.08 0.65
4.67 0.59 0.67 0.65 0.49 0.72 0.56 0.61 0.08 0.62
5.00 0.57 0.65 0.64 0.49 0.69 0.54 0.59 0.08 0.60
5.33 0.56 0.61 0.60 0.43 0.66 0.54 0.57 0.08 0.58
5.67 0.55 0.60 0.58 0.42 0.64 0.51 0.55 0.08 0.57
6.00 0.54 0.57 0.57 0.40 0.62 0.47 0.53 0.08 0.55
6.33 0.53 0.54 0.55 0.37 0.60 0.45 0.51 0.08 0.54
6.67 0.51 0.54 0.49 0.35 0.57 0.43 0.48 0.08 0.50
7.00 0.50 0.51 0.45 0.32 0.55 0.40 0.45 0.09 0.47
7.33 0.46 0.49 0.42 0.30 0.51 0.38 0.43 0.08 0.44
7.67 0.41 0.45 0.38 0.27 0.48 0.36 0.39 0.07 0.40
8.00 0.39 0.41 0.36 0.25 0.44 0.31 0.36 0.07 0.37
Ke [hr '] 0.26 0.27 0.24 0.25 0.22 0.25 0.25 0.02 0.25
tin [hr] 2.69 2.61 2.94 2.74 3.17 2.83 2.83 0.20 2.78
AU C,[(mg/L)- hr] 7.33 7.90 8.22 6.04 8.85 6.77 7.52 1.02 7.62
AUMC, [(mg/L)hr2] 36.59 38.61 38.93 26.64 47.34 32.18 36.72 6.98 37.60
MRT, [hr] 4.99 4.89 4.73 4.41 5.35 4.75 4.85 0.31 4.82
CL [mL/hr/g] 0.68 0.63 0.61 0.83 0.57 0.74 0.68 0.10 0.66





AUMC, [(mg/L)-hr2] 186.06 254.35 156.14 162.96 210.20 638.19 267.98 184.84 198.13
MRToo [hr] 8.91 10.93 8.49 7.98 8.69 15.98 10.16 3.02 8.80
CL [mL/hr/g] 0.48 0.42 0.58 0.46 0.40 0.25 0.43 0.11 0.44


Table 6-10. Unbound muscle voriconazole (10 mg/kg) individual noncompartmental


analysis


pharmacokinetics
Time (hr)


Concentration (mg/L)
102 103 104 105
2.42 2.37 2.80 3.00
2.29 2.25 2.52 2.74
2.00 1.84 2.11 2.22
1.77 1.57 2.09 2.05
1.73 1.54 2.01 2.02
1.69 1.43 1.86 1.96
1.63 1.37 1.81 1.92
1.62 1.35 1.76 1.85
1.47 1.30 1.67 1.82
1.46 1.28 1.57 1.80
1.45 1.26 1.52 1.70
1.37 1.23 1.47 1.68
1.36 1.22 1.41 1.66
1.35 1.21 1.35 1.61
1.32 1.19 1.32 1.55
1.27 1.18 1.27 1.51
1.24 1.16 1.23 1.48
1.22 1.14 1.20 1.46
1.19 1.12 1.16 1.45
1.16 1.09 1.13 1.39
1.13 1.03 1.10 1.32
1.09 1.00 1.06 1.29
1.06 0.98 1.01 1.23
1.03 0.94 0.97 1.19


106 Mean SD Median
3.25 2.74 0.34 2.69
2.91 2.52 0.26 2.47
2.51 2.13 0.23 2.10
2.30 1.97 0.26 2.04
2.19 1.88 0.23 1.91
2.16 1.79 0.26 1.77
2.15 1.74 0.28 1.72
2.10 1.69 0.27 1.69
1.99 1.62 0.26 1.57
1.95 1.58 0.24 1.52
1.91 1.54 0.23 1.49
1.88 1.50 0.24 1.43
1.86 1.47 0.24 1.39
1.77 1.43 0.21 1.35
1.73 1.40 0.20 1.32
1.71 1.36 0.20 1.27
1.64 1.33 0.19 1.23
1.61 1.30 0.19 1.21
1.60 1.28 0.20 1.18
1.58 1.25 0.19 1.16
1.56 1.21 0.20 1.13
1.53 1.18 0.20 1.09
1.49 1.14 0.19 1.06
1.47 1.11 0.20 1.03


0.33
0.67
1.00
1.33
1.67
2.00
2.33
2.67
3.00
3.33
3.67
4.00
4.33
4.67
5.00
5.33
5.67
6.00
6.33
6.67
7.00
7.33
7.67
8.00


Ke [hr-i]
tl/2 [hr]
AUG, [(mg/L)-hr]


0.12 0.09 0.13 0.13 0.12
5.95 7.57 5.49 5.44 5.75
20.88 23.28 18.39 20.43 24.19


0.06 0.11 0.03 0.12
11.30 6.92 2.29 5.85
39.94 24.52 7.84 22.08









Table 6-11. Rat total plasma and unbound muscle voriconazole PK analysis using two-
compartment with non-linear elimination model
Parameter estimates Population estimate (SE%) Between subject variability (BSV) (SE%)
V, (L) 0.50 (1.6) 6.5% (23)
Q (Uhr) 0.28 (8.6) 10.0% (-)
V2 (L) 0.14 (6.5) 18.7% (70.1)
vmax (mg/hr) 0.34 (8.1) 12. 1% (24.6)
Km (mg/L) 1.52 (17.5) 10.0% (-)
fu 0.38 (3.6) 13.0% (41.6)
Residual variability
Total plasma concentration (SE%) Unbound muscle concentration (SE%)
Proportional error 3.1% (21.6) 26.5% (13.5)
Additive error (mg/L)- 0.032 (-)
Note: Dashes indicate that the respective parameters tested during model building were
fixed. % SE: percent standard error of the population parameter estimate.

























I !


SA(1)



V I1m-r Cs
~K. CA"

Figure 6-1. Cascade pharmacokinetic one-compartment model with non-linear elimination of
voriconazole scheme


SA(1)


A(2I)


V IlmLX- Csr
Km + iCA

Figure 6-2. Cascade pharmacokinetic two-compartment model with non-linear elimination of
voriconazole scheme

















10


-4 53

an 55
E -M- ~ 56



a, 1-










0.1
0 2 4 6 8

Time hr

B



4.5


4.0-


S3.51h 55
E -M-t 56
ci 3.0-




c 2.0-
O .






0.5
0 2 4 6 8

Time hr

Figure 6-3. Dosage of 5 mg/kg i.v. bolus: total voriconazole concentration in rat plasma. A)
Linear scale. B) Log scale.

















8





105
9 -M- 106
6-




O
N



2




10

a1






0




0 0
103





110






01






Lier scle B)Lo cae















Representagivq, Pr p otq Dgse=5mg/kg

PRED '---
IPRED -*-



















Time,hr






Rep resentat ve2P plpgts Dqpe= 10mg/kg

PRED '-
IPRED


- 1(o


024168


02468


Time,hr


Figure 6-5. The PK plots using one-compartment with linear elimination model for PK analysis
of rat total plasma voriconazole data: Plot of observed (*), individual predicted
(IPRED) (---) and population predicted (PRED) (---) voriconazole concentration
versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.




































































RES vs PRED


WVRES vs PRED)


-- ~r-'btrkr~a~,


Observed vs predicted concentrations PK~

Population predictions (PRED)


co
E
r
.o



e




ICV
a

o


0 2 4 6 8
Observed concentration, mg/L


Figure 6-6. Goodness of fit plots for observed vs predicted concentrations (* Observed vs
individual predicted, A -Observed vs population predicted) using one-compartment

with linear elimination model for PK analysis of rat total plasma voriconazole
concentration data.


~J1
E
v,
m
~I
o
u,
ao

o
Sm.
II I

po.


_1

E
or
mo
n
(r,
a,,
L~ I


0246810

Population Predicted cI:ncentrationr ng~. 1 I~~jL


0246810

Population Predicted o:ll.ncentratr l r:IInlc L


Figure 6-7. Goodness of fit plots for residuals using one-compartment with linear elimination
model for PK analysis of rat total plasma voriconazole concentration data.














Representative P 5 plot Dgse=5mg/kg

PRED -
IPRED


Representative PYf plpts Dqpe=10mg/kg
~m I PRED
IPRED


O 2 6 8


0 2 4 6


Time,hr


0246810


0246810


Time,hr


Figure 6-8. The PK plots using two-compartment with linear elimination model for PK analysis
of rat total plasma voriconazole data: Plot of observed (*), individual predicted
(IPRED) (---) and population predicted (PRED) (---) voriconazole concentration
versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.











126


























































RES vs PRED


WVRES vs PRED


~t~


Observed vs predicted concentrations PI<

~ i Population predictions (PRED) r
Individual predictions (IPRED)
~I
co
E
r
.o
H~o
a,
o
r
o


o
cu
o_

o


0 2 4 6 8
Observed concentration, mg/L


Figure 6-9. Goodness of fit plots for observed vs predicted concentrations (* Observed vs
individual predicted, A -Observed vs population predicted) using two -compartment
with linear elimination model for PK analysis of rat total plasma voriconazole
concentration data.


O


wt"
E
u,
too
7
rr
v,
1LI


u,
E
v,
LR
m
Is
o
u,
~uo
L11
o
dm
r


0246810

Population Predicted concentration,mglL


0246810

Population Predicted cIncentry;- r ioni~j. Irng:.


Figure 6-10. Goodness of fit plots for residuals using two-compartment with linear elimination
model for PK analysis of rat total plasma voriconazole concentration data.















Representative P~f plotq Dgse=5mg/kg

PRED '-'-
IPRED ~-


Representative2PF~ plpt s D4e=10 mg/kg

PRED -
IPRED -


02(68


02488


Time,hr


024168


02468


Time,hr


Figure 6-11. The PK plots using one-compartment with non-linear elimination model for PK
analysis of rat total plasma voriconazole data: Plot of observed (*), individual
predicted (IPRED) (---) and population predicted (PRED) (---) voriconazole
concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.



128





Observed vs predicted concentrations PK

~ -I Population predictions (PRED) r
Individual predictions (IPRED)

i
co
E
r
.o

r
a,
o
r
o
o
a,
.o
o
acu
a


o


0 2 4 6 8 10

Observed concentration, rng/L


Figure 6-12. Goodness of fit plots for observed vs predicted concentrations (* Observed vs

individual predicted, A-Observed vs population predicted) using one-compartment

with non-linear elimination model for PK analysis of rat total plasma voriconazole
concentration data.


o
r


~nlD
E
u,
nra
=r
n
u,
arm
LT

o


m
E
vl
to
m
=i
n
a;ro

n
Bin
r "


0 2 4 6 8 1

Population Predicted concl~entrallio~nl my1


0246810

Population Predicted coincenirarljon nig.~L


Figure 6-13. Goodness of fit plots for residuals using one-compartment with non-linear

elimination model for PK analysis of rat total plasma voriconazole concentration data.


RES vs PRED


WVRES vs PRED


-rWSLk~i;'~














Representabivq, Pr; plotq Dgse=5mg/kg

PRED '-*-
IPRED --


Representatjve2P s plpts,Dqp~e=10mg/kg

PRED '-'-
IPRED -


012468110
Time,hr


0246110


02468


0246810


Time,hr


Figure 6-14. The PK plots using two-compartment with non-linear elimination model for PK
analysis of rat total plasma voriconazole data: Plot of observed (*), individual
predicted (IPRED) (---) and population predicted (PRED) (---) voriconazole
concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.




















































































_- ----~r


I I I I I I


I I I I I I


Observed vs predicted concentrations PK


s? -Populatinreiton(preits(PRED)7


rs,
a,
E

.n
~co

a,
o
r
o

o
B
o
o
~!CV
a


o


0 2 4 6 8 10

Observed concentration, mg/L


Figure 6-15. Goodness of fit plots for observed vs predicted concentrations (* Observed vs

individual predicted, A-Observed vs population predicted) using two-compartment

with non-linear elimination model for PK analysis of rat total plasma voriconazole

concentration data.


CI



oU,
E
u,
mo
a
v,
a,,
L~I I

o
I


a,
E
cn
LC)
(U
=I
~I
U)
~10

o

II~ I
W


0246810

Population Predicted concent~r~rat~:in mg/~L


0246810

Population Predicted concentration,mglL


Figure 6-16. Goodness of fit plots for residuals using two-compartment with non-linear

elimination model for PK analysis of rat total plasma voriconazole concentration data.


RES vs PRED


WVRES vs PRED





















1.6 .- 0

E. 1.4 _- 0
I 505
0P -M- 506
t 1.2-



S0.8-

N 0.6
o
,9 0.4-
o
0.2

0.0
0 2 4 6 8

Time hr


B



10




505
O -M- 506



o





o



0.1
0 2 4 6 8

Time hr

Figure 6-17. Dosage of 5 mg/kg i.v. bolus: unbound voriconazole concentration in rat muscle. A)
Linear scale. B) Log scale.

















3.5



cn 3.0 .- -10
E -10
1005
,O -N- 1006
2.5-





o
12.0-


0 .5




0 2 4 6 8

Time hr



B





10

-O 1001
S1002
P I-4- 1003





I1-
O




.O




0.1
0 2 4 6 8

Time hr



Figure 6-18. Dosage of 10 mg/kg i.v. bolus: unbound voriconazole concentration in rat muscle.

A) Linear scale. B) Log scale.













Representative Pr; p otq Dgse=5mg/kg

PR ED -
IPRED -


Representative PY plpts,Dqpe=10mg/kg

PRED







1 i I






i I /4


01246110


02468'10
Time,hr


0 2 4 8


0 2 4 6


Time,hr

Figure 6-19. The PK plots using two-compartment with non-linear elimination model for
analysis of rat total plasma (upper curves) and unbound muscle (lower curves)
voriconazole data: Plot of observed (*), individual predicted (IPRED) (---) and
population predicted (PRED) (---) voriconazole concentration versus time. A) 5
mg/kg dose. B) 10 mg/kg dose.

























































RES vs PRED


Observed vs predicted concentrations PK

S2 -1 Population predictions (PRED)
Individual predictions (IPRED)




.O






E0-


0 2 4 6 8 10
Observed concentration, mg/L

Figure 6-20. Goodness of fit plots for observed vs predicted concentrations (* Observed vs
individual predicted, A-Observed vs population predicted) using two-compartment
with non-linear elimination model for PK analysis of rat total plasma and unbound
muscle voriconazole concentration data.


WRESVsPRED
~i
1S~o
E
min
o
a~o
~I
o
~!L"
U]
O
f

0 2 4 6 8 10 12
F~.IF'~ll~it":"~ Predicted o:lnr.tntr~ir;l:llI nig;L


0 2 4 6 8 10 12
Population Predicted o:.ncent~nratioln mgi ~:


Figure 6-21. Goodness of fit plots for residuals using two-compartment with non-linear
elimination model for PK analysis of rat total plasma and unbound muscle
voriconazole concentration data.

















eCptotal
Predicted Cptotal
4 A Cpyree
aCmunbound
Oa -~I Predicted Cmunbound




O
a, 2-








0 2 4 6 8

Time (hr)
B



10
eCptotal
Predicted Cptotal
L I Calculated Cpyree
aCmunbound
OP F Predicted Cmunbound







o





0.1
0 2 4 6 8

Time (hr)

Figure 6-22. Dosage of 5 mg/kg i.v. bolus in rat: average plasma and muscle PK data analysis
using two-compartment with non-linear elimination model. A) Linear scale. B) Log
scale.


















Sprealctea cptotal
8~~ Calculated Cpere
aCmunbound
o -LI Predicted Cmunbound





S4-




0 II


0 2 4 6 8

Time (hr)
B


10













eCptota
-Predicted Cptota
*r A Calculated Cperee
> a Cmunbound
Predicted Cmunbound
0.1
0 2 4 6 8

Time (hr)

Figure 6-23. Dosage of 10 mg/kg i.v. bolus in rat: average plasma and muscle PK data analysis
using two-compartment with non-linear elimination model. A) Linear scale. B) Log
scale.









CHAPTER 7
CONCLUSIONS

The pharmacokinetic and pharmacodynamic characteristics of voriconazole have been

reported in many publications. However, the correlation between in vitro activity and

pharmacokinetic/pharmacodynamic parameters of voriconazole with its clinical outcome in

fungal infections need to be further studied. There is need for mathematical models of this drug

to guide its optimal use.

We did the time-kill and postantifungal- effect (PAFE) experiments for voriconazole

against Candida albicans, Candida glabrata, and Candida parappisiloi isolates. Moreover, a

high-performance liquid chromatography (HPLC) assay was developed to validate these

experiments. Our findings demonstrated that voriconazole exerts prolonged fungistatic activity

against C. albicans, C. glabrata, and C. parapsilosis but no PAFE at concentrations achievable

in human sera with routine dosing. These findings are potentially relevant clinically with other

antifungal agents, to which certain Can2dida isolates exhibit diminished susceptibility or develop

resistance. HPLC confirmed that experiments were conducted at the desired steady-state

voriconazole concentrations. To our knowledge, this is the first study to verify standard time-kill

and PAFE methodologies by directly measuring drug concentrations. These HPLC methods were

essential to the design of dynamic in vitro models to assess the pharmacodynamics of

voriconazole and other agents prior to the achievement of steady-state conditions.

We developed a pharmacokinetic/pharmacodynamic (PK/PD) mathematical model that fits

voriconazole time-kill data against Can2dida isolates in vitro and used the model to simulate the

expected kill curves for typical intravenous and oral dosing regimens. A series of~max

mathematical models were used to fit time-kill data for two isolates each of Can2dida albicans,

Can2dida glabrata and Candida parappisilois PK parameters extracted from human data sets









were used in the model to simulate kill curves for each isolate. Time-kill data were best fit by

using an adapted sigmoidal Emax model that corrected for delays in candidal growth and the onset

of voriconazole activity, saturation of the number of Can2dida and the steepness of the

concentration-response curve. The rates of maximal killing by voriconazole (kmax) were highly

correlated with the growth rates (ks) of the isolates (Pearson' s correlation coefficient = 0.9861).

Simulations using PK parameters derived from the human data sets showed fungistatic effects

against each of the isolates. The developed mathematical PK/PD model linked pharmacokinetic

and pharmacodynamic of voriconazole to provide the basic understanding for defining optimal

antifungal dosing regimens and predict antifungal treatment efficacy.

We established a dynamic system to mimic the in vivo conditions and the PK/PD

relationship of voriconazole against Candida was accurately modeled. Modeling approaches that

utilized human PK data were adapted to define the optimal use of voriconazole and other

antifungal agents.

Microdialysis provided a simple, clean method for determining the free drug concentration

in vitro and in vivo. In order to confirm that the performance of voriconazole microdialysis was

feasible, we conducted the voriconazole protein binding with in vitro microdialysis method. The

pharmacokinetics of voriconazole in vivo and modeling were further investigated. Plasma

concentrations as well as unbound tissue concentrations in rat thigh muscle were measured. A

two-compartment model with non-linear elimination was used to fit both the individual and the

average plasma and muscle voriconazole concentrations and able to produce good curves. The

results from animal study indicated that it is possible to use the free muscle concentrations as a

surrogate marker for the free concentrations in plasma. The free fraction of voriconazole in tissue

(fu) was predicted by PK analysis.









In the past, PK/PD approaches mainly based on the comparisons of pharmacokinetics of

total plasma concentrations and in vitro MIC. Our PK/PD approach is a big improvement over

the previously used. The developed PK/PD modeling approaches enables us to make rational

antifungal agent dosing decisions by predicting the effect of various dosing regimens, taking into

account the clinically effective free concentrations at the target site against candidiasis.









LIST OF REFERENCES


1. Gerzenshtein L, Patel SM, Scarsi KK, Postelnick MJ, Flaherty JP. Breakthrough Candida
infections in patients receiving voriconazole. The Annals of pharmacotherapy. 2005 Jul-
Aug;39(7-8):1342-5.

2. Fratti RA, Belanger PH, Sanati H, Ghannoum MA. The effect of the new triazole,
voriconazole (UK-109,496), on the interactions of Candida albicans and Candida krusei
with endothelial cells. Journal of chemotherapy (Florence, Italy). 1998 Feb; 10(1):7-16.

3. Fukuoka T, Johnston DA, Winslow CA, de Groot MJ, Burt C, Hitchcock CA, et al.
Genetic basis for differential activities of fluconazole and voriconazole against Candida
krusei. Antimicrobial agents and chemotherapy. 2003 Apr;47(4):1213-9.

4. Pfizer Inc. Label: voriconazole for inj section, tablets, oral suspension: LAB-0271-12.
2005 Mar.

5. Boucher HW, Groll AH, Chiou CC, Walsh TJ. Newer systemic antifungal agents :
pharmacokinetics, safety and efficacy. Drugs. 2004;64(18): 1997-2020.

6. Sheehan DJ, Hitchcock CA, Sibley CM. Current and emerging azole antifungal agents.
Clin Microbiol Rev. 1999 Jan; 12(1):40-79.

7. Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial
dosing of mice and men. Clin Infect Dis. 1998 Jan;26(1): 1-10.

8. Theuretzbacher U, Ihle F, Derendorf H. Pharmacokinetic/pharmacodynamic profie of
voriconazole. Clin Pharmacokinet. 2006;45(7):649-63.

9. Pfizer Inc. Label: voriconazole for inj section, tablets, oral suspension: LAB-0271-12.
2005 Mar.

10. Patterson TF. New agents for treatment of invasive aspergillosis. Clin Infect Dis. 2002
Aug 15;35(4):367-9.

11. Sambatakou H, Dupont B, Lode H, Denning DW. Voriconazole treatment for subacute
invasive and chronic pulmonary aspergillosis. The American j journal of medicine. 2006
Jun; 119(6):527 el7-24.

12. Camuset J, Nunes H, Dombret MC, Bergeron A, Henno P, Philippe B, et al. Treatment of
chronic pulmonary aspergillosis by voriconazole in nonimmunocompromi sed patients.
Chest. 2007 May; 13 1(5): 143 5-41.

13. Scott LJ, Simpson D. Voriconazole : a review of its use in the management of invasive
fungal infections. Drugs. 2007;67(2):269-98.

14. Pearson MM, Rogers PD, Cleary JD, Chapman SW. Voriconazole: a new triazole
antifungal agent. The Annals of pharmacotherapy. 2003 Mar;3 7(3):420-32.









15. Espinel-Ingroff A, Boyle K, Sheehan DJ. In vitro antifungal activities of voriconazole
and reference agents as determined by NCCLS methods: review of the literature.
Mycopathologia. 2001;150(3):101-15.

16. Manavathu EK, Ramesh MS, Baskaran I, Ganesan LT, Chandrasekar PH. A comparative
study of the post-antifungal effect (PAFE) of amphotericin B, triazoles and echinocandins
on Aspergillus fumigatus and Candida albicans. The Journal of antimicrobial
chemotherapy. 2004 Feb;53(2):386-9.

17. Lyman CA, Walsh TJ. Systemically administered antifungal agents. A review of their
clinical pharmacology and therapeutic applications. Drugs. 1992 Jul;44(1):9-35.

18. Odds FC, Cheesman SL, Abbott AB. Antifungal effects of fluconazole (UK 49858), a
new triazole antifungal, in vitro. J Antimicrob Chemother. 1986 Oct;18(4):473-8.

19. Como JA, Dismukes WE. Oral azole drugs as systemic antifungal therapy. N Engl J Med.
1994 Jan 27;330(4):263-72.

20. Sanati H, Belanger P, Fratti R, Ghannoum M. A new triazole, voriconazole (UK-
109,496), blocks sterol biosynthesis in Candida albicans and Candida krusei.
Antimicrobial agents and chemotherapy. 1997 Nov;41(11):2492-6.

21. Belanger P, Nast CC, Fratti R, Sanati H, Ghannoum M. Voriconazole (UK-109,496)
inhibits the growth and alters the morphology of fluconazole-susceptible and -resistant
Candida species. Antimicrobial agents and chemotherapy. 1997 Aug;41(8): 1840-2.

22. Li Y, Nguyen MH, Derendorf H, Cheng S, Clancy CJ. Measurement of voriconazole
activity against Candida albicans, C. glabrata, and C. parapsilosis isolates using time-kill
methods validated by high-performance liquid chromatography. Antimicrobial agents and
chemotherapy. 2007 Aug;51(8):2985-7.

23. Espinel-Ingroff A. Mechanisms of resistance to antifungal agents: Yeasts and filamentous
fungi. Rev Iberoam Micol. 2008 Jun;25(2):101-6.

24. Peman J, Salavert M, Canton E, Jarque I, Roma E, Zaragoza R, et al. Voriconazole in the
management of nosocomial invasive fungal infections. Therapeutics and clinical risk
management. 2006 Jun;2(2):129-58.

25. Quindos G, Carrillo-Munoz AJ, Eraso E, Canton E, Peman J. In vitro antifungal activity
of voriconazole: New data after the first years of clinical experience. Rev Iberoam Micol.
2007 Sep 30;24(3):198-208.

26. Espinel-Ingroff A, Arthington-Skaggs B, Iqbal N, Ellis D, Pfaller MA, Messer S, et al.
Multicenter evaluation of a new disk agar diffusion method for susceptibility testing of
filamentous fungi with voriconazole, posaconazole, itraconazole, amphotericin B, and
caspofungin. Journal of clinical microbiology. 2007 Jun;45(6): 181 1-20.









27. Borgers M. Mechanism of action of antifungal drugs, with special reference to the
imidazole derivatives. Rev Infect Dis. 1980 Jul-Aug;2(4): 520-34.

28. Muller FM, Weig M, Peter J, Walsh TJ. Azole cross-resistance to ketoconazole,
fluconazole, itraconazole and voriconazole in clinical Candida albicans isolates from
HIV-infected children with oropharyngeal candidosis. J Antimicrob Chemother. 2000
Aug;46(2):338-40.

29. Cadrobbi J, Hecq J, Lebrun C, Vanbeckbergen D, Jamart J, Galanti L. Long-term
stability of voriconazole 4 mg/mL in dextrose 5% polyvinyl chloride bags at 40C. The
European Journal of Hospital Pharmacy Science. 2006; 12 (3 ):57 9.

30. Langman LJ, Boakye-Agyeman F. Measurement of voriconazole in serum and plasma.
Clin Biochem. 2007 Dec;40(18):1378-85.

31. Roffey SJ, Cole S, Comby P, Gibson D, Jezequel SG, Nedderman AN, et al. The
disposition of voriconazole in mouse, rat, rabbit, guinea pig, dog, and human. Drug
Metab Dispos. 2003 Jun;31(6):731-41.

32. Andes D, Marchillo K, Stamstad T, Conklin R. In vivo pharmacokinetics and
pharmacodynamics of a new triazole, voriconazole, in a murine candidiasis model.
Antimicrobial agents and chemotherapy. 2003 Oct;47(10):3165-9.

33. Sugar AM, Liu XP. Effect of grapefruit juice on serum voriconazole concentrations in the
mouse. Med Mycol. 2000 Jun;38(3):209-12.

34. Saunte DM, Simmel F, Frimodt-Moller N, Stolle LB, Svejgaard EL, Haedersdal M, et al.
In vivo efficacy and pharmacokinetics of voriconazole in an animal model of
dermatophytosis. Antimicrobial agents and chemotherapy. 2007 Sep;51(9):3317-21.

35. Damle B, LaBadie R, Crownover P, Glue P. Pharmacokinetic interactions of efavirenz
and voriconazole in healthy volunteers. British j journal of clinical pharmacology. 2008
Apr;65(4):523-30.

36. Quintard H, Papy E, Massias L, Lasocki S, Arnaud P, Desmonts JM, et al. The
pharmacokinetic profile of voriconazole during continuous high-volume venovenous
hemofiltration in a critically ill patient. Therapeutic drug monitoring. 2008
Feb;30(1):117-9.

37. Jeu L, Piacenti FJ, Lyakhovetskiy AG, Fung HB. Voriconazole. Clin Ther. 2003
May;25(5):1321-81.

38. Sabo JA, Abdel-Rahman SM. Voriconazole: a new triazole antifungal. Ann
Pharmacother. 2000 Sep;34(9): 1032-43.

39. Purkins L, Wood N, Ghahramani P, Greenhalgh K, Allen MJ, Kleinermans D.
Pharmacokinetics and safety of voriconazole following intravenous- to oral-dose
escalation regimens. Antimicrobial agents and chemotherapy. 2002 Aug;46(8):2546-53.









40. Lazarus HM, Blumer JL, Yanovich S, Schlamm H, Romero A. Safety and
pharmacokinetics of oral voriconazole in patients at risk of fungal infection: a dose
escalation study. Journal of clinical pharmacology. 2002 Apr;42(4):395-402.

41. Arikan S, Lozano-Chiu M, Paetznick V, Nangia S, Rex JH. Microdilution susceptibility
testing of amphotericin B, itraconazole, and voriconazole against clinical isolates of
Aspergillus and Fusarium species. J Clin Microbiol. 1999 Dec;37(12):3946-51.

42. Espinel-Ingroff A. In vitro activity of the new triazole voriconazole (UK-109,496)
against opportunistic filamentous and dimorphic fungi and common and emerging yeast
pathogens. J Clin Microbiol. 1998 Jan;36(1):198-202.

43. McGinnis MR, Pasarell L, Sutton DA, Fothergill AW, Cooper CR, Jr., Rinaldi MG. In
vitro activity of voriconazole against selected fungi. Med Mycol. 1998 Aug;36(4):239-
42.

44. Clancy CJ, Nguyen MH. In vitro efficacy and fungicidal activity of voriconazole against
Aspergillus and Fusarium species. Eur J Clin Microbiol Infect Dis. 1998 Aug; 17(8):573-


45. Cordeiro RA, Brilhante RS, Rocha MF, Fechine MA, Costa AK, Camargo ZP, et al. In
vitro activities of caspofungin, amphotericin B and azoles against Coccidioides posadasii
strains from Northeast, Brazil. Mycopathologia. 2006 Jan; 161(1):21-6.

46. Klepser ME, Malone D, Lewis RE, Ernst EJ, Pfaller MA. Evaluation of voriconazole
pharmacodynamics using time-kill methodology. Antimicrobial agents and
chemotherapy. 2000 Jul;44(7):1917-20.

47. Lignell A, Lowdin E, Cars O, Sj olin J. Characterization of the inhibitory effect of
voriconazole on the fungicidal activity of amphotericin B against Candida albicans in an
in vitro kinetic model. The Journal of antimicrobial chemotherapy. 2008 Apr 12.

48. Lozano-Chiu M, Arikan S, Paetznick VL, Anaissie EJ, Rex JH. Optimizing voriconazole
susceptibility testing of Candida: effects of incubation time, endpoint rule, species of
Candida, and level of fluconazole susceptibility. J Clin Microbiol. 1999 Sep;37(9):2755-


49. Nguyen MH, Yu CY. Voriconazole against fluconazole-susceptible and resistant candida
isolates: in-vitro efficacy compared with that of itraconazole and ketoconazole. J
Antimicrob Chemother. 1998 Aug;42(2):253-6.

50. Johnson EM, Szekely A, Warnock DW. In-vitro activity of voriconazole, itraconazole
and amphotericin B against filamentous fungi. J Antimicrob Chemother. 1998
Dec;42(6):741-5.

51. Dannaoui E, Meletiadis J, Mouton JW, Meis JF, Verweij PE. In vitro susceptibilities of
zygomycetes to conventional and new antifungals. J Antimicrob Chemother. 2003
Jan;51(1):45-52.









52. Varkey JB, Perfect JR. Rare and emerging fungal pulmonary infections. Seminars in
respiratory and critical care medicine. 2008 Apr;29(2): 121-31.

53. Alanio A, Brethon B, Feuilhade de Chauvin M, de Kerviler E, Leblanc T, Lacroix C, et
al. Invasive pulmonary infection due to Trichoderma longibrachiatum mimicking
invasive Aspergillosis in a neutropenic patient successfully treated with voriconazole
combined with caspofungin. Clin Infect Dis. 2008 May 15;46(10):ell6-8.

54. Martin MV, Yates J, Hitchcock CA. Comparison of voriconazole (UK-109,496) and
itraconazole in prevention and treatment of Aspergillus fumigatus endocarditis in guinea
pigs. Antimicrobial agents and chemotherapy. 1997 Jan;41(1): 13-6.

55. Kirkpatrick WR, McAtee RK, Fothergill AW, Rinaldi MG, Patterson TF. Efficacy of
voriconazole in a guinea pig model of disseminated invasive aspergillosis. Antimicrobial
agents and chemotherapy. 2000 Oct;44(10):2865-8.

56. Capilla J, Guarro J. Correlation between in vitro susceptibility of Scedosporium
apiospermum to voriconazole and in vivo outcome of scedosporiosis in guinea pigs.
Antimicrobial agents and chemotherapy. 2004 Oct;48(10):4009-11.

57. Lionakis MS, Lewis RE, May GS, Wiederhold NP, Albert ND, Halder G, et al. Toll-
deficient Drosophila flies as a fast, high-throughput model for the study of antifungal
drug efficacy against invasive aspergillosis and Aspergillus virulence. J Infect Dis. 2005
Apr 1;191(7):1188-95.

58. Weerasuriya N, Snape J. Oesophageal candidiasis in elderly patients: risk factors,
prevention and management. Drugs & aging. 2008;25(2): 119-30.

59. Heinz WJ, Einsele H. Caspofungin for treatment of invasive aspergillus infections.
Mycoses. 2008;51 Suppl 1:47-57.

60. Proia LA, Tenorio AR. Successful use of voriconazole for treatment of Coccidioides
meningitis. Antimicrobial agents and chemotherapy. 2004 Jun;48(6):2341.

61. Stanzani M, Vianelli N, Bandini G, Paolini S, Arpinati M, Bonifazi F, et al. Successful
treatment of disseminated Fusariosis after allogeneic hematopoietic stem cell
transplantation with the combination of voriconazole and liposomal amphotericin B. The
Journal of infection. 2006 Dec;53(6):e243-6.

62. Nesky MA, McDougal EC, Peacock Jr JE. Pseudallescheria boydii brain abscess
successfully treated with voriconazole and surgical drainage: case report and literature
review of central nervous system pseudallescheriasis. Clin Infect Dis. 2000
Sep;31(3):673-7.

63. Kappe R. Antifungal activity of the new azole UK-109, 496 (voriconazole). Mycoses.
1999;42 Suppl 2:83-6.









64. Ally R, Schurmann D, Kreisel W, Carosi G, Aguirrebengoa K, Dupont B, et al. A
randomized, double-blind, double-dummy, multicenter trial of voriconazole and
fluconazole in the treatment of esophageal candidiasis in immunocompromised patients.
Clin Infect Dis. 2001 Nov 1;33(9):1447-54.

65. Kullberg BJ, Sobel JD, Ruhnke M, Pappas PG, Viscoli C, Rex JH, et al. Voriconazole
versus a regimen of amphotericin B followed by fluconazole for candidaemia in non-
neutropenic patients: a randomised non-inferiority trial. Lancet. 2005 Oct 22-
28;366(9495):1435-42.

66. Hwang YK, Joo NH, Tee GY, Tan P, Tan L, Girij a R. Clinical activity of the new
triazole drug voriconazole (UK 109, 496) against disseminated hepatosplenic
aspergillosis in a patient with relapsed leukemia. Haematologia (Budap). 2001;31(1):73-
80.

67. Herbrecht R. Voriconazole: therapeutic review of a new azole antifungal. Expert review
of anti-infective therapy. 2004 Aug;2(4):485-97.

68. Singh N, Limaye AP, Forrest G, Safdar N, Munoz P, Pursell K, et al. Combination of
voriconazole and caspofungin as primary therapy for invasive aspergillosis in solid organ
transplant recipients: a prospective, multicenter, observational study. Transplantation.
2006 Feb 15;81(3):320-6.

69. de la Pena A, Liu P, Derendorf H. Microdialysis in peripheral tissues. Adv Drug Deliv
Rev. 2000 Dec 15;45(2-3):189-216.

70. Yokoyama M, Suzuki E, Sato T, Maruta S, Watanabe S, Miyaoka H. Amygdalic levels of
dopamine and serotonin rise upon exposure to conditioned fear stress without elevation of
glutamate. Neurosci Lett. 2005 Apr 29;379(1):37-41.

71. Humpel C, Ebendal T, Olson L. Microdialysis: a way to study in vivo release of
neurotrophic bioactivity: a critical summary. J Mol Med. 1996 Sep;74(9):523-6.

72. Lee SC, Fung CP, Huang JS, Tsai CJ, Chen KS, Chen HY, et al. Clinical correlates of
antifungal macrodilution susceptibility test results for non-AIDS patients with severe
Candida infections treated with fluconazole. Antimicrobial agents and chemotherapy.
2000 Oct;44(10):2715-8.

73. Liu P, Fuhrherr R, Webb AI, Obermann B, Derendorf H. Tissue penetration of
cefpodoxime into the skeletal muscle and lung in rats. Eur J Pharm Sci. 2005 Jul-
Aug;25(4-5):439-44.

74. Liu P, Muller M, Grant M, Obermann B, Derendorf H. Tissue penetration of
cefpodoxime and cefixime in healthy subjects. Journal of clinical pharmacology. 2005
May;45(5):564-9.









75. Traunmuller F, Zeitlinger M, Zeleny P, Muller M, Joukhadar C. Pharmacokinetics of
single- and multiple-dose oral clarithromycin in soft tissues determined by microdialysis.
Antimicrobial agents and chemotherapy. 2007 Sep;51(9):3185-9.

76. Hartings JA, Gugliotta M, Gilman C, Strong AJ, Tortella FC, Bullock MR. Repetitive
cortical spreading depolarizations in a case of severe brain trauma. Neurological research.
2008 Jun 3.

77. Hutschala D, Skhirtladze K, Kinstner C, Mayer-Helm B, Muller M, Wolner E, et al. In
vivo microdialysis to measure antibiotic penetration into soft tissue during cardiac
surgery. The Annals of thoracic surgery. 2007 Nov;84(5): 1605-10.

78. Tyvold SS, Solligard E, Lyng O, Steinshamn SL, Gunnes S, Aadahl P. Continuous
monitoring of the bronchial epithelial lining fluid by microdialysis. Respiratory research.
2007;8:78.

79. Muller M, Brunner M, Schmid R, Mader RM, Bockenheimer J, Steger GG, et al.
Interstitial methotrexate kinetics in primary breast cancer lesions. Cancer research. 1998
Jul 15;58(14):2982-5.

80. Hansen DK, Davies MI, Lunte SM, Lunte CE. Pharmacokinetic and metabolism studies
using microdialysis sampling. J Pharm Sci. 1999 Jan;88(1):14-27.

81. de Lange, Elizabeth C. M., de Boer, A. G., Breimer DD. Methodological issues in
microdialysis sampling for pharmacokinetic studies. Advanced Drug Delivery Reviews.
2000 2000/12/15;45(2-3):125-48.

82. Parent M, Bush D, Rauw G, Master S, Vaccarino F, Baker G. Analysis of Amino Acids
and Catecholamines, 5-Hydroxytryptamine and Their Metabolites in Brain Areas in the
Rat Using in Vivo Microdialysis. Methods. 2001 2001/1;23(1):11-20.

83. Day JC, Kornecook TJ, Quirion R. Application of in vivo microdialysis to the study of
cholinergic systems. Methods. 2001 Jan;23(1):21-39.

84. Hows ME, Lacroix L, Heidbreder C, Organ AJ, Shah AJ. High-performance liquid
chromatography/tandem mass spectrometric assay for the simultaneous measurement of
dopamine, norepinephrine, 5-hydroxytryptamine and cocaine in biological samples. J
Neurosci Methods. 2004 Sep 30; 13 8(1-2): 123-32.

85. Baseski HM, Watson CJ, Cellar NA, Shackman JG, Kennedy RT. Capillary liquid
chromatography with MS3 for the determination of enkephalins in microdialysis samples
from the striatum of anesthetized and freely-moving rats. J Mass Spectrom. 2005
Feb;40(2):146-53.

86. Bowser MT, Kennedy RT. In vivo monitoring of amine neurotransmitters using
microdialysis with on-line capillary electrophoresis. Electrophoresis. 2001
Oct;22(17):3668-76.









87. Huynh BH, Fogarty BA, Martin RS, Lunte SM. On-line coupling of microdialysis
sampling with microchip-based capillary electrophoresis. Anal Chem. 2004 Nov
1;76(21):6440-7.

88. Kanamori K, Ross BD. Suppression of glial glutamine release to the extracellular fluid
studied in vivo by NMR and microdialysis in hyperammonemic rat brain. J Neurochem.
2005 Jul;94(1):74-85.

89. Lucas LH, Wilson SF, Lunte CE, Larive CK. Concentration profiling in rat tissue by
high-resolution magic-angle spinning NMR spectroscopy: investigation of a model drug.
Anal Chem. 2005 May 1;77(9):2978-84.

90. Mayer BX, Namiranian K, Dehghanyar P, Stroh R, Mascher H, Muller M. Comparison of
UV and tandem mass spectrometric detection for the high-performance liquid
chromatographic determination of diclofenac in microdialysis samples. J Pharm Biomed
Anal. 2003 Nov 24;33(4):745-54.

91. Kanamori K, Kondrat RW, Ross BD. 13C enrichment of extracellular neurotransmitter
glutamate in rat brain--combined mass spectrometry and NMR studies of
neurotransmitter turnover and uptake into glia in vivo. Cell Mol Biol (Noisy-le-grand).
2003 Jul;49(5):819-36.

92. Lindon JC. HPLC-NMR-MS: past, present and future. Drug Discov Today. 2003 Nov
15;8(22):1021-2.

93. Lindon JC, Nicholson JK, Wilson ID. Directly coupled HPLC-NMR and HPLC-NMR-
MS in pharmaceutical research and development. J Chromatogr B Biomed Sci Appl.
2000 Oct 1;748(1):233-58.

94. Lommen A, Godej ohann M, Venema DP, Hollman PC, Spraul M. Application of directly
coupled HPLC-NMR-MS to the identification and confirmation of quercetin glycosides
and phloretin glycosides in apple peel. Anal Chem. 2000 Apr 15;72(8):1793-7.

95. Li Y, Peris J, Zhong L, Derendorf H. Microdialysis as a tool in local pharmacodynamics.
The AAPS j journal. 2006;8(2):E222-3 5.

96. Weikop P, Egestad B, Kehr J. Application of triple-probe microdialysis for fast
pharmacokinetic/pharmacodynamic evaluation of dopamimetic activity of drug
candidates in the rat brain. Journal of neuroscience methods. 2004 Dec 30; 140(1-2):59-
65.

97. Feng MR, Turluck D, Burleigh J, Lister R, Fan C, Middlebrook A, et al. Brain
microdialysis and PK/PD correlation of pregabalin in rats. European j ournal of drug
metabolism and pharmacokinetics. 2001 Jan-Jun;26(1-2):123 -8.

98. Au-Yeung SC, Riggs KW, Gruber N, Rurak DW. The use of microdialysis for the study
of drug kinetics: central nervous system pharmacokinetics of diphenhydramine in fetal,









newborn, and adult sheep. Drug metabolism and disposition: the biological fate of
chemicals. 2007 Aug;35(8):1285-91.

99. Schuck VJ, Rinas I, Derendorf H. In vitro microdialysis sampling of docetaxel. Journal of
pharmaceutical and biomedical analysis. 2004 Nov 19;36(4):807-13.

100. Derendorf H, Lesko LJ, Chaikin P, Colbum WA, Lee P, Miller R, et al.
Pharmacokinetic/pharmacodynamic modeling in drug research and development. Journal
of clinical pharmacology. 2000 Dec;40(12 Pt 2): 13 99-418.

101. Meibohm B, Derendorf H. Basic concepts of pharmacokinetic/pharmacodynamic
(PK/PD) modelling. Intemational joumal of clinical pharmacology and therapeutics.
1997 Oct;35(10):401-13.

102. Li Y, Nguyen MH, Cheng S, Schmidt S, Zhong L, Derendorf H, et al. A
pharmacokinetic/pharmacodynamic mathematical model accurately describes the activity
of voriconazole against Candida spp. in vitro. International j journal of antimicrobial
agents. 2008 Apr;31(4):369-74.

103. Xu J, Winkler J, Derendorf H. A pharmacokinetic/pharmacodynamic approach to predict
total prednisolone concentrations in human plasma. Journal of pharmacokinetics and
pharmacodynamics. 2007 Jun;34(3):355-72.

104. Derendorf H, Hochhaus G, Krishnaswami S, Meibohm B, Mollmann H. Optimized
therapeutic ratio of inhaled corticosteroids using retrometabolism. Die Pharmazie. 2000
Mar;55(3):223-7.

105. Derendorf H, Meibohm B. Modeling of pharmacokinetic/pharmacodynamic (PK/PD)
relationships: concepts and perspectives. Pharmaceutical research. 1999 Feb; 16(2): 176-
85.

106. Walsh TJ, Karlsson MO, Driscoll T, Arguedas AG, Adamson P, Saez-Llorens X, et al.
Pharmacokinetics and safety of intravenous voriconazole in children after single- or
multiple-dose administration. Antimicrobial agents and chemotherapy. 2004
Jun;48(6):2166-72.

107. Louie A, Liu QF, Drusano GL, Liu W, Mayers M, Anaissie E, et al. Pharmacokinetic
studies of fluconazole in rabbits characterizing doses which achieve peak levels in serum
and area under the concentration-time curve values which mimic those of high-dose
fluconazole in humans. Antimicrobial agents and chemotherapy. 1998 Jun;42(6):1512-4.

108. Ohtani Y, Kotegawa T, Tsutsumi K, Morimoto T, Hirose Y, Nakano S. Effect of
fluconazole on the pharmacokinetics and pharmacodynamics of oral and rectal
bromazepam: an application of electroencephalography as the pharmacodynamic method.
Journal of clinical pharmacology. 2002 Feb;42(2): 183-91.









109. Louie A, Drusano GL, Banerj ee P, Liu QF, Liu W, Kaw P, et al. Pharmacodynamics of
fluconazole in a murine model of systemic candidiasis. Antimicrobial agents and
chemotherapy. 1998 May;42(5): 1105-9.

110. Katayama M, Igarashi H, Tani K, Nezu Y, Harada Y, Yogo T, et al. Effect of multiple
oral dosing of fluconazole on the pharmacokinetics of cyclosporine in healthy beagles.
The Journal of veterinary medical science / the Japanese Society of Veterinary Science.
2008 Jan;70(1):85-8.

111. Carrasco-Portugal Mdel C, Flores-Murrieta FJ. Gender differences in the oral
pharmacokinetics of fluconazole. Clinical drug investigation. 2007;27(12):851-5.

112. Saari TI, Laine K, Neuvonen M, Neuvonen PJ, Olkkola KT. Effect of voriconazole and
fluconazole on the pharmacokinetics of intravenous fentanyl. European j journal of clinical
pharmacology. 2008 Jan;64(1):25-30.

113. Andes D, Marchillo K, Stamstad T, Conklin R. In vivo pharmacodynamics of a new
triazole, ravuconazole, in a murine candidiasis model. Antimicrobial agents and
chemotherapy. 2003 Apr;47(4):1193-9.

114. Andes D, Marchillo K, Conklin R, Krishna G, Ezzet F, Cacciapuoti A, et al.
Pharmacodynamics of a new triazole, posaconazole, in a murine model of disseminated
candidiasis. Antimicrobial agents and chemotherapy. 2004 Jan;48(1):137-42.

115. Greer ND. Posaconazole (Noxafil): a new triazole antifungal agent. Proc (Bayl Univ Med
Cent). 2007 Apr;20(2):188-96.

116. Courtney R, Pai S, Laughlin M, Lim J, Batra V. Pharmacokinetics, safety, and tolerability
of oral posaconazole administered in single and multiple doses in healthy adults.
Antimicrob Agents Chemother. 2003 Sep;47(9):2788-95.

117. Janssen-Pharmaceutica. Product information: SPORANOX@ (ITRACONAZOLE)
ORAL SUSPENSION: Hepatotoxicity Labeling -062602. 2002 June.

118. Pfizer. Product information: DIFLUCAN@ (Fluconazole): LAB-0099 -6. 1. 2004 August.

119. Clancy CJ, Huang H, Cheng S, Derendorf H, Nguyen MH. Characterizing the effects of
caspofungin on Candida albicans, Candida parapsilosis, and Candida glabrata isolates by
simultaneous time-kill and postantifungal-effect experiments. Antimicrobial agents and
chemotherapy. 2006 Jul;50(7):2569-72.

120. Gage R, Stopher DA. A rapid HPLC assay for voriconazole in human plasma. J Pharm
Biomed Anal. 1998 Sep;17(8):1449-53.

121. MacCallum DM, Whyte JA, Odds FC. Efficacy of caspofungin and voriconazole
combinations in experimental aspergillosis. Antimicrob Agents Chemother. 2005
Sep;49(9):3697-701.









122. Pfaller MA, Diekema DJ, Rex JH, Espinel-Ingroff A, Johnson EM, Andes D, et al.
Correlation of MIC with outcome for Candida species tested against voriconazole:
analysis and proposal for interpretive breakpoints. Journal of clinical microbiology. 2006
Mar; 44(3):819-26.

123. NCCLS. Reference method for broth dilution antifungal susceptibility testing of yeasts.
Approved standard, 2nd ed. NCCLS document M27-A2. National Committee for Clinical
Laboratory Standards, Wayne, PA.; 2002.

124. Smith J, Safdar N, Knasinski V, Simmons W, Bhavnani SM, Ambrose PG, et al.
Voriconazole therapeutic drug monitoring. Antimicrobial agents and chemotherapy. 2006
Apr;50(4): 1570-2.

125. Liu P, Rand KH, Obermann B, Derendorf H. Pharmacokinetic-pharmacodynamic
modelling of antibacterial activity of cefpodoxime and cefixime in in vitro kinetic
models. International journal of antimicrobial agents. 2005 Feb;25(2): 120-9.

126. Schuck EL, Derendorf H. Pharmacokinetic/pharmacodynamic evaluation of anti-infective
agents. Expert review of anti-infective therapy. 2005 Jun;3(3):361-73.

127. Treyaprasert W, Schmidt S, Rand KH, Suvanakoot U, Derendorf H.
Pharmacokinetic/pharmacodynamic modeling of in vitro activity of azithromycin against
four different bacterial strains. International journal of antimicrobial agents. 2007
Mar;29(3):263-70.

128. Canton E, Peman J, Gobernado M, Viudes A, Espinel-Ingroff A. Synergistic activities of
fluconazole and voriconazole with terbinafine against four Candida species determined
by checkerboard, time-kill, and Etest methods. Antimicrobial agents and chemotherapy.
2005 Apr;49(4):1593-6.

129. Manavathu EK, Cutright JL, Chandrasekar PH. Organism-dependent fungicidal activities
of azoles. Antimicrobial agents and chemotherapy. 1998 Nov;42(1 1):3018-21.

130. Simonsen L, Jorgensen A, Benfeldt E, Groth L. Differentiated in vivo skin penetration of
salicylic compounds in hairless rats measured by cutaneous microdialysis. Eur J Pharm
Sci. 2004 Feb;21(2-3):379-88.

131. Sasongko L, Williams KM, Day RO, McLachlan AJ. Human subcutaneous tissue
distribution of fluconazole: comparison of microdialysis and suction blister techniques.
British j journal of clinical pharmacology. 2003 Nov;56(5):5 51-61.

132. Fair W, Jr. An Algorithm for Weighted Linear Regression. Available from:
http://www.codeproj ect. com/KB/recipes/LinReg .aspx

133. Arauj o BV, Silva CF, Haas SE, Dalla Costa T. Microdialysis as a tool to determine free
kidney levels of voriconazole in rodents: A model to study the technique feasibility for a
moderately lipophilic drug. Journal of pharmaceutical and biomedical analysis. 2008 Feb
29.









134. Kusbeci T, Avci B, Cetinkaya Z, Ozturk F, Yavas G, Ermis SS, et al. The effects of
caspofungin and voriconazole in experimental Candida endophthalmitis. Current eye
research. 2007 Jan;32(1):57-64.

135. Breit SM, Hariprasad SM, Mieler WF, Shah GK, Mills MD, Grand MG. Management of
endogenous fungal endophthalmitis with voriconazole and caspofungin. American
journal of ophthalmology. 2005 Jan; 139(1): 13 5-40.

136. Verweij PE, de Pauw BE, Meis JF. Voriconazole (Pfizer ltd). IDrugs. 1999 Sep;2(9):925-
37.

137. Yang H, Wang Q, Elmquist WF. The design and validation of a novel intravenous
microdialysis probe: application to fluconazole pharmacokinetics in the freely-moving rat
model. Pharmaceutical research. 1997 Oct;14(10): 1455-60.

138. Tsai TH. Assaying protein unbound drugs using microdialysis techniques. Journal of
chromatography. 2003 Nov 25;797(1-2): 161-73.

139. Liu Z, Yang XG, Li X, Pan W, Li J. Study on the ocular pharmacokinetics of ion-
activated in situ gelling ophthalmic delivery system for gatifloxacin by microdialysis.
Drug development and industrial pharmacy. 2007 Dec;33(12): 1327-31.

140. Uehara T, Sumiyoshi T, Itoh H, Kurata K. Lactate production and neurotransmitters;
evidence from microdialysis studies. Pharmacology, biochemistry, and behavior. 2008
Aug;90(2):273-81.

141. Sumiyoshi T, Tsunoda M, Uehara T, Tanaka K, Itoh H, Sumiyoshi C, et al. Enhanced
locomotor activity in rats with excitotoxic lesions of the entorhinal cortex, a
neurodevelopmental animal model of schizophrenia: behavioral and in vivo microdialysis
studies. Neuroscience letters. 2004 Jul 1;364(2):124-9.

142. Apparaju SK, Gudelsky GA, Desai PB. Pharmacokinetics of gemcitabine in tumor and
non-tumor extracellular fluid of brain: an in vivo assessment in rats employing
intracerebral microdialysis. Cancer chemotherapy and pharmacology. 2008
Feb;61(2):223-9.

143. Zhou Q, Gallo JM. In vivo microdialysis for PK and PD studies of anticancer drugs. The
AAPS journal. 2005;7(3):E659-67.

144. Schwarz AJ, Zocchi A, Reese T, Gozzi A, Garzotti M, Varnier G, et al. Concurrent
pharmacological MRI and in situ microdialysis of cocaine reveal a complex relationship
between the central hemodynamic response and local dopamine concentration.
Neurolmage. 2004 Sep;23(1):296-304.

145. Egashira N, Ishigami N, Mishima K, Iwasaki K, Oishi R, Fujiwara M. Delta9-
Tetrahydrocannabinol-induced cognitive deficits are reversed by olanzapine but not
haloperidol in rats. Progress in neuro-psychopharmacology & biological psychiatry. 2008
Feb 15;32(2):499-506.









146. van Duuren E, van der Plasse G, van der Blom R, Joosten RN, Mulder AB, Pennartz CM,
et al. Pharmacological manipulation of neuronal ensemble activity by reverse
microdialysis in freely moving rats: a comparative study of the effects of tetrodotoxin,
lidocaine, and muscimol. The Joumnal of pharmacology and experimental therapeutics.
2007 Oct;323(1):61-9.

147. Nakashima K, Yamamoto K, Al-Dirbashi OY, Kaddoumi A, Nakashima MN. Semi-
micro column HPLC of triazolam in rat plasma and brain microdialysate and its
application to drug interaction study with itraconazole. Journal of pharmaceutical and
biomedical analysis. 2003 Jan 15;30(6):1809-16.

148. Mathy FX, Ntivunwa D, Verbeeck RK, Preat V. Fluconazole distribution in rat dermis
following intravenous and topical application: a microdialysis study. Journal of
pharmaceutical sciences. 2005 Apr;94(4):770-80.

149. Kondoh S, Nagasawa S, Kawanishi M, Yamaguchi K, Kajimoto S, Ohta T. Effects of
ebselen on cerebral ischemia and reperfusion evaluated by microdialysis. Neurological
research. 1999 Oct;21(7):682-6.

150. Stimpfel TM, Gershey EL. Selecting anesthetic agents for human safety and animal
recovery surgery. Faseb J. 1991 Apr;5(7):2099-104.

151. Martenson M, Houts J, Heinricher M, Ogden B. A simple device for humidification of
inspired gases during volatile anesthesia in rats. Contemporary topics in laboratory
animal science / American Association for Laboratory Animal Science. 2005
Mar;44(2):46-8.

152. Yang R, Tan X, Kenney RJ, Jr., Thomas A, Landis M, Qureshi N, et al. Hemorrhagic
shock in the rat: comparison of carotid and femoral cannulation. The Journal of surgical
research. 2008 Jan; 144(1): 124-6.

153. Beaumont DM, Mark TM, Jr., Hills R, Dixon P, Veit B, Garrett N. The effects of chrysin,
a Passiflora incarnata extract, on natural killer cell activity in male Sprague-Dawley rats
undergoing abdominal surgery. AANA journal. 2008 Apr;76(2): 113-7.

154. Arauj o BV, Conrado DJ, Palma EC, Dalla Costa T. Validation of rapid and simple LC-
MS/MS method for determination of voriconazole in rat plasma. Journal of
pharmaceutical and biomedical analysis. 2007 Aug 15;44(4):985-90.

155. Muller M, dela Pena A, Derendorf H. Issues in pharmacokinetics and pharmacodynamics
of anti-infective agents: distribution in tissue. Antimicrobial agents and chemotherapy.
2004 May;48(5):1441-53.

156. Tegeder I, Lotsch J, Kinzig-Schippers M, Sorgel F, Kelm GR, Meller ST, et al.
Comparison of tissue concentrations after intramuscular and topical administration of
ketoprofen. Pharmaceutical research. 2001 Jul;18(7):980-6.









157. Muller M, Rohde B, Kovar A, Georgopoulos A, Eichler HG, Derendorf H. Relationship
between serum and free interstitial concentrations of cefodizime and cefpirome in muscle
and subcutaneous adipose tissue of healthy volunteers measured by microdialysis.
Journal of clinical pharmacology. 1997 Dec;37(12): 1108-13.









BIOGRAPHICAL SKETCH

Yanjun Li was born in 1974, in Hunan, China. She got her M.D. in Hunan Medical

University (current name: Central South University Xiangya School of Medicine). She started

her Ph.D. program in August 2004 in the Department of Pharmaceutics, College of Pharmacy,

University of Florida, under the supervision of Dr. Hartmut Derendorf. Yanjun received her

Ph.D. in pharmaceutics in August 2008.





PAGE 1

1 PHARMACOKINETICS/PHARMACODYNAMICS OF VORICONAZOLE By YANJUN LI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Yanjun Li

PAGE 3

3 To my beloved husband, daughters, and parents, for their love

PAGE 4

4 ACKNOWLEDGMENTS I express m y deep appreciation and grateful thanks to Dr. Hartmut Derendorf for his intelligent guidance and generous support throughout my Ph.D. study. He gave me the opportunity of being part of his group explor ing the opportunities in research area of pharmacokinetics and pharmacodynamics. The knowle dge I learned from him benefits me and will be cherished in my entire lifetime. Thanks go to Dr. Cornelius J. Clancy and Dr. Minh-Hong Nguyen for their broad knowledge in microbiology and infectious diseases intelligent guidance a nd generous support in the lab space and equipment necessary for my e xperiments. It has really been a wonderful experience to learn knowledge in their laborat ories. Without their help the pharmacodynamic studies would not have been performed. I tha nk Dr. Veronika Butterweck for her support and guidance during the animal studies and in my project s. I offer my grateful thanks to the members of my supervisory committee, Dr. Guenther Ho chhaus and Dr. Kenneth Rand for their support and valuable advice. I thank Stephan, Jian, Vipul, Sab for th e study discussion and help. I thank the administrative people of the Department of Pharmaceutics, Mr. Marty Rhoden, Mrs. Patricia Khan, and Mrs. Robin Keirnan-Sanchez, for their techni cal support. I also would like to extend my thanks to all faculties of the Departm ent of Pharmaceutics, staffs, graduate students, and post-doc fellows who were my colleagues for their support and friendship.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.......................................................................................................................10 ABSTRACT...................................................................................................................................13 CHAP TER 1 INTRODUCTION..................................................................................................................15 Background.............................................................................................................................15 Voriconazole...........................................................................................................................15 Mechanisms of Action..................................................................................................... 16 Drug Stability..................................................................................................................17 Animal Pharmacokinetic Studies.................................................................................... 17 Human Pharmacokinetic Studies..................................................................................... 17 In Vitro and In Vivo Pharm acodynamic Studies..............................................................18 Clinical Efficacy..............................................................................................................20 Microdialysis..........................................................................................................................21 Principle of Microdialysis...............................................................................................22 Features of Microdialysis................................................................................................ 22 Application in PK/PD Studies......................................................................................... 23 The PK/PD Approaches.......................................................................................................... 24 Hypothesis and Objectives..................................................................................................... 25 2 MEASUREMENT OF VORICONAZOLE AC TIVITY AGAINST CANDIDA ISOLATES USING TIME-KILL M ETHODS VALIDATED BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY............................................................ 30 Background.............................................................................................................................30 Specific Aims..........................................................................................................................30 Materials and Methods...........................................................................................................30 Antifungal Agents........................................................................................................... 30 Test Isolates.....................................................................................................................30 Antifungal Susceptibility Testing.................................................................................... 31 Antifungal Carryover....................................................................................................... 31 Time-kill Expe rim ents..................................................................................................... 32 Postantifungal Effect (PAFE) Experiments..................................................................... 32 HPLC for Determination of Voriconazole in S tock Solution and RPMI Medium......... 32 Calibration Curves for Voriconazole in Stock S olution and RPMI Medium.................. 33 Statistical Analysis.......................................................................................................... 33 Results.....................................................................................................................................34

PAGE 6

6 Stability of Voriconazole in Stock S olution and RPMI Medium.................................... 34 Minimum Inhibitory Concentration (MIC)..................................................................... 34 Time-kills and Postantifungal Effect (PAFE)................................................................. 34 Discussion...............................................................................................................................36 3 APPLYING PHARMACOKINETIC/PHARM ACODYNAMIC MATHEMATICAL MODEL ACCURATELY DESCRIBES TH E ACTIVITY OF VORICONAZOLE AGAINST CANDIDA SPP. IN VITRO ..................................................................................50 Background.............................................................................................................................50 Specific Aim...........................................................................................................................51 Materials and Methods...........................................................................................................51 Mathematical Modelling of Timekill Data.................................................................... 51 Simulations of Expected Timek ill Curves Using Hum an PK Data............................... 51 Statistical Analysis.......................................................................................................... 52 Results.....................................................................................................................................52 An Adapted Sigmoidal Emax Model Provides the Best F it for Voriconazole Time kill Data against Candida Isolates................................................................................ 52 Human PK Data for Voriconazole Can Be Used to Sim ulate Expected Timekill........ 54 Discussion...............................................................................................................................55 4 PHARMACODYNAMIC STUDY IN DYNAM I C SYSTEM FOR DESCRIBING THE ACTIVITY OF VORICONAZOLE AGAINST CANDIDA SPP. IN VITRO ........................62 Background.............................................................................................................................62 Specific Aim...........................................................................................................................62 Materials and Methods...........................................................................................................63 Antifungal Agents........................................................................................................... 63 Test Isolates.....................................................................................................................63 Softwares.........................................................................................................................63 Dynamic Model Design................................................................................................... 63 Time-kill Experiments in the Dynamic Model................................................................ 65 Results.....................................................................................................................................66 Discussion...............................................................................................................................69 5 IN VITRO MICRODIALYSIS OF VORICONAZOLE........................................................81 Background.............................................................................................................................81 Specific Aims..........................................................................................................................81 Materials and Methods...........................................................................................................82 Materials..........................................................................................................................82 Methods...........................................................................................................................82 Recovery...................................................................................................................82 Extraction efficiency method (EE)...........................................................................82 Retrodialysis method (RD)....................................................................................... 83 Microdialysis experiments....................................................................................... 84 HPLC for determination of voriconazole................................................................. 84

PAGE 7

7 Calibration curves for voriconazole......................................................................... 85 Results.....................................................................................................................................85 Stability of Voriconazole in L actate Ringers Solution and Plasm a............................... 86 Protein Binding................................................................................................................86 Discussion...............................................................................................................................86 6 IN VIVO PHARMACOKINETIC STUDIES OF VORICONAZOLE IN RATS ................. 91 Background.............................................................................................................................91 Specific Aims..........................................................................................................................91 Materials and Methods...........................................................................................................92 Reagents and Equipment................................................................................................. 92 Animals............................................................................................................................92 Experimental Design....................................................................................................... 92 Anesthetic Procedure.......................................................................................................93 Antiseptic Procedure.......................................................................................................93 Blood Samples.................................................................................................................93 Microdialysis...................................................................................................................94 Probe calibration...................................................................................................... 94 Muscle microdialysis................................................................................................ 95 Data Analysis...................................................................................................................95 Noncompartmental pharm acokinetic analysis ......................................................... 96 Compartmental pharmacokineti c analysis and modeling .........................................97 Results...................................................................................................................................102 Probe Recovery............................................................................................................. 102 Individual Pharmacokinetic Analysis of Total Voricona zole in Plasma after Intravenous Bolus Administration............................................................................. 103 Noncompartmental pharm acokinetic analysis ....................................................... 103 Compartmental pharmacokinetic analysis............................................................. 103 One-compartment model with linear elim ination analysis..................................... 104 Two-compartment model with li near elim ination analysis.................................... 104 One-compartment model with nonlinear elimination............................................ 105 Two-compartment model with nonlinear elim ination............................................ 105 Individual Pharmacokinetic Analysis of Unbound Voriconazole in Muscle after Intravenous Bolus Adm inistration............................................................................. 106 Noncompartmental pharm acokinetic analysis ....................................................... 107 Compartmental pharmacokinetic analysis............................................................. 107 Two-compartment model with nonlinear elim ination............................................ 107 Pharmacokinetic Analysis of Average Total Voriconazole Data in Plasma and Unbound Voriconazole Data in Muscle af ter Intravenous Bolus Adm inistration..... 108 Discussion.............................................................................................................................109 7 CONCLUSIONS.................................................................................................................. 138 LIST OF REFERENCES.............................................................................................................141 BIOGRAPHICAL SKETCH.......................................................................................................155

PAGE 8

8 LIST OF TABLES Table page 1-1 Pharmacokinetic parameters of voriconazole in mouse, rat, rabbit, guinea pig and dog following single an d multiple administration by oral and intravenous routes............ 26 1-2 Pharmacokinetic (PK) prope rties of triazole antifungals ................................................... 27 2-1 Percentage (%) of voriconazole maintenan ce in stock solution and RPMI m edium at different temperatures........................................................................................................38 2-2 Minimum inhibitory concen tration (MIC) of voriconazole against Candida isolates ....... 38 2-3 Candida albicans ATCC90029 in constant conc entration experim ents............................ 39 2-4 Candida albicans SC5314 in constant concentration experim ents....................................39 2-5 Candida glabrata 1 in constant concen tration experim ents..............................................40 2-6 Candida glabrata 2 in constant concen tration experim ents..............................................40 2-7 Candida parapsilosis 1 in constant concentration experim ents........................................41 2-8 Candida parapsilosis 2 in constant concentration experim ents........................................41 2-9 Summarized voriconazole time-k ill d ata against Candida isolates.................................... 42 2-10 Candida albicans ATCC90029 in PAFE experiments ......................................................42 2-11 Candida glabrata 2 in P AFE experiments......................................................................... 43 2-12 Candida glabrata 1 in P AFE experiments......................................................................... 43 2-13 Candida parapsilosis 1 in PAFE experim ents................................................................... 44 2-14 Candida parapsilosis 2 in PAFE experim ents................................................................... 44 3-1 Pharmacodynamic parameters and goodness of fit criteria agains t Candida isolates ........ 57 3-2 Steady-state pharmacokinetic paramete rs in plasm a as calculated by twocompartment model analysis.............................................................................................. 58 4-1 Candida albicans ATCC90029 in changing concentration experim ents.......................... 71 4-2 Candida glabrata 1 in changing concentration experim ents............................................. 72 4-3 Candida glabrata 2 in changing concentration experim ents............................................. 73

PAGE 9

9 4-4 Candida parapsilosis 1 in cha nging concentration experim ents........................................74 4-5 Candida parapsilosis 2 in cha nging concentration experim ents........................................75 4-6 Pharmacodynamic parameters and goodness of fit criteria against Candida isolates in the dynam ic infection model.............................................................................................. 76 5-1 List of materials.......................................................................................................... .......88 5-2 Extraction efficiency (EE) method for m easuring voriconazole recovery........................ 89 5-3 Retrodialysis (RD) method for m easuring voriconazole recovery.................................... 89 5-4 Rat plasma protein binding da ta of voriconazole m easured by in vitro microdialysis...... 90 5-5 Human plasma protein binding da ta of voriconazole m easured by in vitro microdialysis......................................................................................................................90 6-1 Recovery data of voriconazole using re trodialysis (RD) m ethod in rats muscle............. 112 6-2 Total plasma voriconazole (5 mg/kg) individual noncom partmental PK analysis.......... 113 6-3 Total plasma voriconazole (10 mg/kg) individual noncom partmental PK analysis........ 114 6-4 Comparison of objective function value (OFV) and Akaike inform ation criterion (AIC) value from rats total plasma voricona zole PK analysis using different models.... 115 6-5 One-compartment model with linear elimin ation f or rat total plasma voriconazole PK analysis.............................................................................................................................116 6-6 Two-compartment model with linear elim ination f or rat total plasma voriconazole PK analysis.......................................................................................................................116 6-7 One-compartment with non-linear elim ination m odel for rat total plasma voriconazole PK analysis.................................................................................................117 6-8 Two-compartment with non-linear elimination model for rat total plasm a voriconazole PK analysis.................................................................................................117 6-9 Unbound muscle voriconazole (5 mg /kg) individual noncompartm ental pharmacokinetics analysis............................................................................................... 118 6-10 Unbound muscle voriconazole (10 mg /kg) individual noncom partmental pharmacokinetics analysis............................................................................................... 119 6-11 Rat total plasma and unbound muscle voriconazole PK analysis using twocom partment with non-linear elimination model............................................................. 120

PAGE 10

10 LIST OF FIGURES Figure page 1-1 Structural relati onship among azole drugs ......................................................................... 28 1-2 Plasma voriconazole concen tr ation-time profiles following i.v dosing and following oral dosingof voriconazole.................................................................................................29 2-1 Susceptibility of voriconazo le against Candida isolates ....................................................45 2-2 Culture flasks for voriconazo le against Candida isolates in v itro ......................................45 2-3 Time-kill curves for voriconazo le ag ainst Candida isolates.............................................. 46 2-4 Voriconazole did not demonstrate PAFEs......................................................................... 47 2-5 Chromatogram of voriconazole which con centration was determ ined from peak area..... 48 2-6 Voriconazole concentrations in cultu re m edia throughout the duration of time-kill experiments by HPLC........................................................................................................48 2-7 Voriconazole concentrations in PAFE experiments by HPLC.......................................... 49 3-1 Plasma VOR concentration-time profile s sim ulated with a two-compartment PK model..................................................................................................................................59 3-2 Fitted time-kill curves derived using the m athematical model for constant concentrations of voriconazole..........................................................................................60 3-3 Simulations of candidal timekills and plasma voriconazole concentrationtime profiles. ..............................................................................................................................61 4-1 Concentration elimination curve of voric onazole in the dynam ic infection model........... 77 4-2 Time-kill curves from the dynamic in vitro m odel............................................................ 78 4-3 Fitted time-kill curves were derived by our m athematical model for changing concentrations of voriconazole..........................................................................................79 4-4 Using parameters from dynamic models to simulate candidal time-kills and plasma voriconazole concentration-tim e profiles.......................................................................... 80 6-1 Cascade pharmacokinetic one-compartment m odel with non-linear elimination of voriconazole scheme........................................................................................................121 6-2 Cascade pharmacokinetic two-compartmen t m odel with non-linear elimination of voriconazole scheme........................................................................................................121

PAGE 11

11 6-3 Dosage of 5 mg/kg i.v. bolus: total voriconazole concentration in rat plasm a................122 6-4 Dosage of 10 mg/kg i.v. bolus: total voriconazole c oncentration in rat plasm a..............123 6-5 The PK plots using one-compartment with linear elim ination model for PK analysis of rat total plasma voriconazole data............................................................................... 124 6-6 Goodness of fit plots for observed vs predicted concentrations using onecom partment with linear elimination mode l for PK analysis of rat total plasma voriconazole concentration data...................................................................................... 125 6-7 Goodness of fit plots for residuals using one-compartm ent with linear elimination model for PK analysis of rat total pl asma voriconazole concentration data.................... 125 6-8 The PK plots using two-compartment with linear elim ination model for PK analysis of rat total plasma voriconazole data............................................................................... 126 6-9 Goodness of fit plots for observed vs predicted concentrations using two com partment with linear elimination mode l for PK analysis of rat total plasma voriconazole concentration data...................................................................................... 127 6-10 Goodness of fit plots for residuals using two-compartm ent with linear elimination model for PK analysis of rat total pl asm a voriconazole concentration data.................... 127 6-11 The PK plots using one-compartment w ith non-linear elim ination model for PK analysis of rat total plasma voriconazole data................................................................. 128 6-12 Goodness of fit plots for observed vs predicted concentrations using onecom partment with non-linear elimination mode l for PK analysis of rat total plasma voriconazole concentration data...................................................................................... 129 6-13 Goodness of fit plots for residuals using one-compartm ent with non-linear elimination model for PK analysis of ra t total plasma voriconazole concentration data...................................................................................................................................129 6-14 The PK plots using two-compartment w ith non-linear elim ination model for PK analysis of rat total plasma voriconazole data................................................................. 130 6-15 Goodness of fit plots for observed vs predicted concentr ations using twocom partment with non-linear elimination mode l for PK analysis of rat total plasma voriconazole concentration data...................................................................................... 131 6-16 Goodness of fit plots for residuals using two-compartm ent with non-linear elimination model for PK analysis of ra t total plasma voriconazole concentration data...................................................................................................................................131 6-17 Dosage of 5 mg/kg i.v. bolus: unbound voriconazole c oncentration in rat muscle ......... 132

PAGE 12

12 6-18 Dosage of 10 mg/kg i.v. bolus: unbound voriconazole conc entration in rat m uscle....... 133 6-19 The PK plots using two-compartment with non-linear elim ination model for analysis of rat total plasma and unbound muscle voriconazole data............................................. 134 6-20 Goodness of fit plots for observed vs predicted concentr ations using twocom partment with non-linear elimination mode l for PK analysis of rat total plasma and unbound muscle voriconazole concentration data....................................................135 6-21 Goodness of fit plots for residuals using two-compartm ent with non-linear elimination model for PK analysis of rat total plasma and unbound muscle voriconazole concentration data...................................................................................... 135 6-22 Dosage of 5 mg/kg i.v. bol us in rat: average plasma and muscle PK data analysis using two-compartment with non-linear elimination model............................................ 136 6-23 Dosage of 10 mg/kg i.v. bolus in rat: average p lasma and muscle PK data analysis using two-compartment with non-linear elimination model............................................ 137

PAGE 13

13 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PHARMACOKINETICS/PHARMACODYNAMICS OF VORICONAZOLE By Yanjun Li August 2008 Chair: Hartmut Derendorf Major: Pharmaceutical Sciences Candida spp. is the most common cause of opportunistic mycoses worldwide, especially the invasive opportunistic mycotic infection. Oropharyngeal ca ndidiasis is the most common opportunistic infection associat ed with AIDS. Candidemia is the fourth most common bloodstream infection in th e developed world and is associated with mortality rates of up to 38% despite treatment with available antifungals. Voriconazole a triazole agent that is effective in treating candidiasis and aspergillo sis, inhibits ergosterol synt hesis by blocking the action of 14 demethylase. It has also lowered toxicity compared with previous antifungal agents ; however, it is used to cure candidiasis inside the hum an body after oral administration or i.v. injection. Empiric antifungal therapy is initiated as the first measure dealing with fungal infection, and there are no suitable mathematical models of voriconazole antifungal activity esta blished for rational guidance to maximize efficien cy and minimize toxicity. The aim of these studies was to devel op an optimal pharmacokinetic/pharmacodynamic (PK/PD) model to predict clinical outcome and improve the effectivenes s of voriconazole. The PK/PD model integrates the pharmacokinetics of unbound voriconazole with in vitro antifungal activity against certain Candida strains for rational dosing of this agent.

PAGE 14

14 In these studies, we measured voriconazole activity against Candida isolates using timekill methods validated by high performance liquid chromatography in vitro and a pharmacokinetic/pharmacodynamic mathematical model based on time-kill curves against different Candida strains was used to accurately describe the antifungal activity of voriconazole. Using this model, pharmacodynamic studies in dy namic system for describing the activity of voriconazole against Candida spp. were conducted. Moreover, in vitro microdialysis was performed to determine protein binding of voric onazole in both rat and human plasma. Also, in vivo microdialysis in Wistar rats after intrave nous administration of voriconazole was conducted to evaluate unbound muscle conc entrations of voriconazole an d to compare the free tissue concentration to free plasma concentration. We investigated the voriconazole PK profile by analyzing the total concentra tions in plasma and unbound concentrations in muscle and developed a population PK model to fit the vor iconazole PK data in rats using NONMEM. These studies provided new insights to improve the drug clinical efficacy by using the developed voriconazole PK/PD model.

PAGE 15

15 CHAPTER 1 INTRODUCTION Background In the past two decades, the incidence of invasive fungal infections, especially in immunocompromised patients, such as AIDS patie nts have risen significantly. Invasive fungal infections are growing in fre quency and optimal selection and dosing of antifungal agents are important. Agents administered for invasive infect ions are amphotericin B, flucytosine, and azole antifungals. Azoles, a class of organic compound s having a five-membered heterocyclic ring with two double bonds, have gained attention as useful drugs with antifungal activity in vitro against a broad spectrum of fungal pathogens. The azole antifungal agents inhibit the synthesis of ergosterol by blocking the action of cytochrome P450 14a-demethylase (P45014DM) [1-4]. The structure of six main azole drugs includi ng ketoconazole, fluconazole, itraconazole, voriconazole, ravuconazole, and posaconazole is shown in Figure 1-1 [5, 6]. A ntimicrobial PD describes the relationship between drug exposure and treatment efficacy. Therapeutic outcome predictions based upon these PD relationships ha ve correlated well in treatment against both susceptible and resistant pathogens The potential value of using PK/PD parameters as guides for establishing optimal dosing regimens for new and old drugs, for new emerging pathogens and resistant organisms, for setting susceptibility breakpoints, and for reducing the cost of drug development should make the continuing search for the therapeutic rationa le of antifungal dosing of animals and humans worthwhile [1, 7]. Voriconazole Voriconazole (UK-109,496) is a triazole that is structurally related to fluconazole (Figure 1-1) [5, 8]. It is developed by Pfizer Pharmaceuticals (http://www.pfizer.com) and its clinical use was approved by the Food and Drug Administra tion (FDA) in May 2002. The trade name of

PAGE 16

16 voriconazole is Vfend [9]. Voriconazole recently has gained increasing attention as a new class of azole antifungal drug with its lower toxi city and expanded antifungal activity against Candida spp. compared with previous antifungal therapies, and also with its a dvantage of high activity against Aspergillus species [10-12]. Voriconazole is fungistatic and exhibits no PAFE against Candida albicans [13-15]. Time-kill and PAFE data are limited against C. glabrata and do not exist against C. parapsilosis isolates. Moreover, standard time-kill and PAFE methodologies, although widely used, have not been validated fo r voriconazole or other antifungals by direct measurements of drug concentrations [16]. Mechanisms of Action All azole antifungal agents work principall y by inhibition of P45014DM. This enzyme is in the sterol biosynthesi s pathway that leads from lanosterol to ergosterol [6, 17-19]. Compared to fluconazole, voriconazole inhibits P45014DM to a greater extend. This inhibition is dosedependent [20-22]. Voriconazole also has an enhanced antifungal spectrum that includes filamentous fungi [23-26]. Voriconazole also inhibits 24-methylene dihydrolanasterol demethylation in certain yeast and filamentous fungi. Other antifungal effects of azole compounds have been proposed and include: inhibition of endogenous respiration, interaction with membrane phospholipids, and inhibition of th e transformation of yeasts to mycelial forms [19]. Other mechanisms may involve inhibition of purine uptake and impairment of triglyceride and/or phospholipid biosynthesis [27]. Cross-resistance of voriconazole with other azole antifungals has been demonstrated, proba bly due to common modes of action [28]. Due to the potential for cross-resistance, specific organism susceptibility data shou ld be reviewed before selecting an antifungal for the treatment of infections.

PAGE 17

17 Drug Stability The stability of voriconazole has been tested under a variety of conditions. The voriconazole chemical and physical in-use stab ility has been demonstrated for 24 hours at 2 to 8C (36 to 46F) following reconstitution of the lyophile with water [4]. Based on a shelf-life of 90% residual potency, voriconazole in 5% dextrose solutions were stable for at least 15 days at 4C [29]. It was repoted that the stabili ty of voriconazole in serums and plasma was found to be stable for up to 7 days at room te mperature, for 30 days frozen at 20 C, and through 3 freeze thaw cycles [30]. Voriconazole diluted in 0.9% Sodium Chloride or Lactated Ringers solution can be stable for 14 days at 15 to 30C [4]. Animal Pharmacokinetic Studies Voriconazoles pharm acokinetics and metabolism have been studied in mouse, rat, rabbit, dog, guinea pig, and humans after single and multiple administration by both oral and intravenous routes [31-36]. Its pharmacokinetics parameters ha ve been shown in Table 1-1. Oral absorption of voriconazole is high (~75%) in anim als, including rats, mice, rabbits, guinea pigs, and dogs [31, 37]. Drug distribution studie s using radiolabeled 14C-voriconazole 10 mg/kg in male and female rats showed extensive distribution th roughout tissues. In preclin ical studies in rats, voriconazole was excreted in both urine and feces, with the majority of drug eliminated within 48 hours of administration [37]. Human Pharmacokinetic Studies Voriconazole is orally and pa renterally active. B ioavailability is up to 96% with peak plasma levels occurring 1-2 hours after dosing. Pl asma protein binding is roughly 58% and is not affected by renal or hepatic disease. The volume of distribution is 4.6L/k g, indicating widespread distribution in the body [37, 38]. With administration of the reco mmended intravenous (IV) or oral loading dose, steady-state concentr ations are reached within 24 hours. Without the loading dose,

PAGE 18

18 accumulation must occur for 6 days to reach steady state. Oral steady state plasma concentrations have ranged from 2.1 to 4.8 mg/L (peak) and 1.4 to 1.8 mg/L (trough). Mean Cmax and area under the curve (AUC) are reduced by 34% and 24%, respectively, following a high fat meal. The absorption of voriconazole is not a ffected by increases in gastric pH [37, 38]. Voriconazole displays nonlinear pharmacokinetics due to saturation of its metabolism. Increasing the IV dose from 3 mg/kg to 4 mg/kg twice daily and the oral dose from 200 mg to 300 mg twice daily results in roughly a 2.5 fold increase in the AUC [39, 40]. Mean plasma voriconazo le concentration-time profiles after i.v. (day 7) and oral (day 14) admini stration are illustrated in Figure 1-2. Cmax occurred at the end of the 1h i.v. infusion and between 1.4 and 1.8 h after oral administration [39]. Human data indicate voriconazo le concentrations in the C SF are between 40% and 70% of the concentrations in the plasma [37, 38]. Voriconazole is a substrate for, is extensively metabolized by, and is an inhibitor of cytochrome P450 enzymes 2C19, 2C9 and 3A4. Enzyme CYP2C19 exhibits genetic polymorphism resu lting in an approximately 4-fold higher voriconazole exposure in poor metabolizers vs. extensive metabolizers. Eight metabolites have been identified, of which three are major N-oxide metabolites. The N-oxide metabolites do not exhibit antifungal activity. Voriconazole metabo lites are primarily excreted renally. Roughly 85% of a dose appears in the urine with < 2% as unchanged drug. The elimination half life has been reported as 6 hours, but in patients rece iving prolonged therapy up to 6 days have been required to recover 90% of th e drug in the urine and feces [37, 38]. An overview of pharmacokinetic properties of voriconazole and othe r triazoles from clinical studies is provided in Table 1-2. In Vitro and In Vivo Pharmacodynamic Studies Pharm acodynamics refers to the time course a nd intensity of drug effects on the organism, whether human or experimental animal. To desc ribe antifungal efficacy, parameters such as

PAGE 19

19 minimum inhibitory concentrati on (MIC), time above the MIC, time-kill curves, sub-MIC effects and post-antifungal effects (PAFEs) are used. In vitro and in vivo studies are conducted to examine the effects of the drug on antimicrobial inhibition or killing, th e rate and extent of killing over time, and the durati on of antimicrobial effects. Voriconazole has broad in vitro antifungal activity against a variety of fungi including Candida spp., Aspergillus spp., Cryptococcus neoformans, Blastomyces dermatitidis, Coccidioides immitis, Histoplasma capsulatum, Fusarium spp., and Penicillium marneffei [41-45]. The voriconazole pharmacodynamics was evaluated using time-kill methods. It reported that voriconazole had nonconcentration-dependent activity in vitro with maximum effect observed at 3 times the MIC [46]. Voriconazole is a fungistatic agent against Candida spp. and Cryptococcus neoformans but a fungicidal drug against Aspergillus spp. [44]. Voriconazole has shown fungicidal activity in vitro against multiple As pergillus spp., including A. terreus which is inherently amphoteric in Bresistant [5]. Although voriconazole MICs for fluconazole-resistant isolates were significantly higher than those for fluconazole-susceptibl e isolates, it enhances activity against fluconazole-resistant Candi da krusei, Candida glabrata, and Candida guilliermondii [5, 6, 47]. Some isolates which are resistant to fluconazole and/or itraconazole may, expectedly, exhibit cross -resistance to voriconazole [48, 49]. Voriconazole has no activity against the agents of Zygomycetes, such as Muco r spp. and Rhizomucor spp. which generate considerably high voriconazole MICs [6, 50, 51]. Voriconazole has demonstrated anti-fungi act ivity in a number of animal models of systemic Candidiasis, pulmonary, dissemina ted and intravascular fungal infection [5, 6, 52, 53]. Moreover, it was found that serum voriconazole concentrations were very low and often undetectable in mice, which was due to a co mbination of high clearance and extensive

PAGE 20

20 metabolism by cytochrome P450 enzymes. Thus, mice were abandoned as being suitable for further study of voriconazole and most subsequent work with this drug has been performed in the guinea pig [33]. In an early study, it has been shown that in neutropenic guinea pigs with systemic Candidiasis, voriconazole was more active than fluc onazole or itraconazole in animals infected with C. krusei, C. glabrata, or az ole-resistant strains of C. albicans [6]. A study has also been shown that voriconazole is highl y efficacious in both the preventi on and treatment of Aspergillus endocarditis in the guinea pig and is supe rior to itraconazole in these respects [54]. In 2000, voriconazole was evaluated in an immunosuppressedguinea pig model of invasive Aspergillosis. The study indicated that voriconazole was more eff ective than amphotericin B or similar doses of itraconazole and improved survival and si gnificantly reduced tissue colony counts [55]. Recently, some investigators have evaluated the efficacy of voriconazole in a systemic infection by Scedosporium apiospermum in immunodepressed guinea pigs and documented that voriconazole prolonged survival and reduced fungal load in ki dney and brain tissues of the animals infected with the first strain but was unabl e to prolong survival or to redu ce fungal load in brain tissue for the latter strain [56]. Voriconazole also demonstrated activit y against Aspergillosis in the rat and rabbit animal models, although the pharmacokine tics of voriconazole in these animals are suboptimal [6]. Moreover, the Drosophila fly was also developed as a fast, high-throughput model to study the drug efficacy against Aspergillosis and Aspergillus [57]. The study documented that Toll-deficient Drosophila flies treated by voriconazo le, had significantly bette r survival rates and lower tissue fungal burdens than control. Clinical Efficacy Voriconazole has been proven as a prom isi ng agent for the treatment of a number of invasive fungal infections incl uding Candidiasis, acute and chr onic invasive aspergillosis, coccidioidomycosis, cryptococcosis, fusarios is, scedosporiosis and pseudallescheriasis,

PAGE 21

21 paecilomycosis and other endemic mycoses in clinical studies [5, 13, 58-63]. In a randomized, double-blind, double-dummy, multicenter trial of vor iconazole and fluconazole in the treatment of esophageal candidiasis in immunocompromise d patients, it proved that voriconazole was as effective as fluconazole in the treatment of biopsy-proven esophageal candidiasis in immunocompromised patients [64]. More recently, a study have shown that voriconazole was as effective as the regimen of amphotericin B followed by fluconazole in the treatment of candidemia in non-neutropenic patient s, and with fewer toxic effects [65]. Interestingly, voriconazole has shown good response against di sseminated hepatosplenic aspergillosis in a patient with relapsed leukemia, whereas itraconazole, amphotericin B and liposomal amphotericin have no efficacy [66]. In Phase II/III trials, voriconazole was well-tolerated and demonstrated extremely clinical efficacy in pa tients with fluconazole-sensitive and -resistant candida infection, aspergillosis, and various refractory fungal infections [67]. The efficacy of the combination of voriconazole and other anti-fungi drug such as caspofungin, have also studied in the therapy for invasive aspergillosis in organ transplant. It documented that combination of voriconazole and caspofungin might be considered preferable therapy for subsets of organ transplant recipients with invasive aspergillosis, such as those with renal failure or A. fumigatus infection [68]. Microdialysis In m any cases the clinical outcome of th erapy needs to be determined by the drug concentration in the tissue compartment in which the pharmacological effect occurs rather than in the plasma. Microdialysis (MD) is an in vivo technique which allows direct measurement of unbound tissue concentrations and permits monito ring of the biochemical and physiological effects of drugs on the body th roughout the body. MD is a cath eter-based sampling technique that provides the opportunity to sample analytes from interstitial fluid from tissues to measure

PAGE 22

22 the free pharmacologically active co ncentration of exogenous or e ndogenous compounds at a site closer to the target site than plasma [28]. The method has been used extensively in animal and human studies for decades [28, 69-72]. Principle of Microdialysis Microd ialysis is a sampling technique that is used to measure the concentration of the unbound fraction of endogenous and/or exogenous substances in the extr acellular fluid of many tissues (e.g. adipose tissue, brain, heart, lung or solid tumors) [28, 73-79]. It is applicable to both animal and human studies. The basic principle is to mimic the function of a capillary blood vessel by perfusing a thin dialysis probe with physiological fluid af ter it is inserted in the tissue of interest by means of a guide cannula. Conti nuous transfer of soluble molecules from the extracellular fluid into the probe occurs by means of a semipermeable membrane covering the tip of the probe. Samples are subsequently collected either at intermittent time points for later analysis by standard chemical an alytical techniques, or more r ecently, continuously for direct on-line analysis [28, 80-83]. Features of Microdialysis There are many advantages of m icrodialysis [28, 80-83]. It is a straight-forward technique that is not too difficult to establish on a routine basis. Microdialysis sampling does not change the net fluid balance of the surrounding sample matrix a nd provides clean samples in which analytes are separated from further enzymatic action. Because there is no net fluid loss, samples can be collected continuously for hours or days from a single freely moving animal. Most importantly, each animal can serve as its own control, redu cing the number of required subjects compared with some other methods such as direct tissue assay. It can be used in humans in a relatively noninvasive manner. Moreover, microdialysis is easily coupled with other chemical analysis, such as high performance liquid chromatography (HPLC) [83, 84], mass spectrometry (MS) [84, 85], capillary

PAGE 23

23 electrophoresis (CE) [86, 87], or nuclear magnetic resonance (NMR) [88, 89]. These techniques are often combined, e.g., HPLC-MS [84, 90], HPLC-NMR [89, 91], HPLC-NMR-MS [92-94] and CE-LIF [86] to enhance the specificity, sens itivity, reliability and efficien cy of separation and detection. As with any technique, there are limitati ons in the applicatio n of microdialysis [80-83, 95-97]. Microdialysis requires sensitive analytical methods to measure low concentrations in small sample volumes. Implantation of the probe almost certainly leads to tissue reactions that can interfere with the physiologica l system under investigation. To minimize this interference, determination of optimal times after probe implan tation must be determined specifically for the analyte of interest. For example, the optimal time to measure endogenous dopamine levels after probe insertion may not be the most optimal time to measure glutamate levels due to the continuing process of gliosis at the site of pr obe tissue damage. Another problem is associated with highly lipophilic drugs stic king to tubing and probe compone nts, thereby complicating the relationship between dialysate a nd extracellular concentrations. Most importantly, microdialysis causes dilution of analyte levels in two ways. First, endogenous an alyte levels may be decreased near the probe leading to a tissue reaction and change in physiological status. Second, the diluting effect of the dialysis procedure leads to lower concentrations of analyte in samples compared to tissue, requiring both sensitive analytical methods as well as the need to determine in vivo recovery of the analyte to calculate true concentrations in the extracellular fluid. This latter technique may impose tem poral limitations (e.g. the no-net-flux method may require hours of sampling) and may be confounded by physiolo gical conditions that change over time. Application in PK/PD Studies MD has opened m any possibilities to study PK and PD processes in different tissues [74, 98, 99]. Its many features make this technique interest ing to be applied in PK and PD studies. MD provides a clean sample, which is protein free. Because of the membrane porosity and cut-off,

PAGE 24

24 large molecules such as proteins cannot get thro ugh the membrane, which allows determining the free drug concentration in the tissue [100, 101]. It is very important when only the free drug concentration is considered to be pharmacologically active. The PK/PD Approaches Pharm acokinetics (PK) describes and predicts the time course of drug concentrations and pharmacodynamics (PD) refers to the time course and intensity of drug effects [102-105]. PK/PD models integrate the PK and PD a pproach, in which the variable of time is incorporated into the relationship of effect to concentration [8, 102, 106]. For voriconazole, inte grating the PK and PD approaches provides information about the relationships between in vitro susceptibility, dosage, drug concentrations in the body a nd antifungal or toxicological eff ects, which helps developers select a rational dosage regimen for confirmatory clinical testing [106-109]. PK/PD issues have been studied most extensively for fluconazole [109-112]. A study documented that the PD characteristic of flucon azole most closely associated with outcome was the ratio of the area under the concentration-time curve from 0 to 24 h to the MIC (24 h AUC:MIC) in a murine model of candidemia [109]. In a neutropenic murine model of disseminated Candida albicans in fection, the PK/PD parameters for voriconazole, ravuconazole and posaconazole were also similarly characte rized, which support that the 24h AUC/MIC ratio is the critical PK/PD parameter a ssociated with treatment efficacy [32, 113, 114]. However, an optimal mathematical model of voriconazoles antifungal effect has not been established yet. It is important to establish a mathematical model to predict clinical outcome of voriconazole combining PK data from in vivo m odel studies and PD data from in vitro model studies. In conclusion, the studies of PK and PD profile of voriconazole play an important role in its clinical use. Simulation th rough PK/PD modeling is also an important tool to capture the

PAGE 25

25 variability and uncertainty that is implicitly inhere nt in the azole antifungal drugs further development. Hypothesis and Objectives The overall hypothesis is that an optim al math ematical model of voriconazoles antifungal effect could be established to predict clinical outcome of them combining pharmacokinetic data from in vivo model studies and pharmacodynamics data from in vitro model for more rational use to improve the effectiveness of the drug therap eutics in clinic. To test the hypothesis of this study, the following specific aims were purposed: Specific Aim 1: Measure voriconazole activity against Candida isolates using time-kill methods validated by high performance liquid chromatography. Specific Aim 2: Apply pharmacokinetic/pharmacodynamic mathematical model to accurately describe the activ ity of voriconazole against candida spp. in vitro Specific Aim 3: Conduct pharmacodynamic studies in dynamic system for describing the activity of voriconazole against Candida spp. Specific aim 4: Determine the value of protein bindi ng of voriconazole in both rat and human plasma using in vitro microdialysis. Specific aim 5: Investigate the voriconazole PK profile from analyzing the total voriconazole concentrations in plasma and free concentrations in muscle in rats by in vivo microdialysis. Develop a populat ion PK model of voriconazo le in rats using NONMEM.

PAGE 26

26 Table 1-1. Pharmacokinetic parameters of voriconazole in mouse, rat, rabbit, guinea pig and dog following single and multiple administra tion by oral and intravenous routes [31] Parameter MouseRat RabbitGuinea pigDog Sex MaleMaleFemaleFemale FemaleaMale/Female Number of animalsb3223 1c4 Plasma protein binding(%)676666604551 Intravenous Dose (mg/kg)1010103103 Single dose AUCt (gh/ml)41.718.681.61.138.532.1 Multiple dose AUCt (gh/ml)d8.06.713.91.62217.9 Oral Dose (mg/kg)30303010106 Single dose Cmax(g/ml) 12.49.516.714.1 6.5 Single dose Tmax (gh/ml)2611 8 3 Single dose AUCt (gh/ml)98.890215.63.22988.8 Multiple dose AUCt (gh/ml)d35.332.357.44.432.352.2 Apparent bioavailability(%)81159888775138 a Male animals were used for single oral dose study. b Number of animals per time point, except for dog and rabbit studies, wh ich involved serial bleeding. c n=3 per time for single oral dose study in guinea pigs. d Once daily for up to 10 days (minimum 5 days), except guinea pig multiple oral (three times daily).

PAGE 27

27 Table 1-2. Pharmacokinetic pr operties of triazole antifungals [8, 115-118] Property Fluconazole Itraconazole Voriconazole Posaconazole Bioavailability >90% 50%% >95% Variable (depending on dosage regime and food) Protein binding 11% 99% 58% >98% Volume of distribution(L/kg) 0.7.8 11 4.6 7 Tmax (h) 2 4 1 3 CL (L/h/kg) 0.014 0.2-0.4 0.2-0.5 0.2-0.5 Metabolism Hepatic: 11% metabolized Hepatic: CYP3A4 Hepatic: CYP2C19, 2C9, 3A4 Hepatic: glucuronidation to inactive metabolites Elimination halflife 22 hours 35 hours 6 hours (variable) 15 hours Elimination route 80% excreted unchanged in urine Hepatic; <1% excreted unchanged in urine Hepatic; <2% excreted unchanged in urine <1% excreted unchanged in urine; 66% excreted unchanged in feces

PAGE 28

28 Figure 1-1. Structural rela tionship among azole drugs [5, 6]

PAGE 29

29 Figure 1-2. Plasma voriconazole con centration-time profiles following i.v dosing (day 7) and following oral dosing (day 14) of voriconazole [39]

PAGE 30

30 CHAPTER 2 MEASUREMENT OF VORICONAZOLE AC TIVITY AGAINST CANDIDA ISOLATES USING TIME-KILL M ETHODS VALIDATED BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY Background Voriconazole is a triazole agent that i nhibits ergosterol synthesis by blocking the action of 14 -demethylase. The drug is fungistatic and exhibits no postantifungal effect (PAFE) against Candida albicans (1, 3, 7, 9). Time-kill and PAFE data for Candida glabrata are limited (1, 7) and do not exist for Candida parapsilosi s isolates. Moreover, standard time-kill and PAFE methodologies although widely used, have not been validated for voriconazole or other an tifungals by direct measurement of drug concentrations. Specific Aims This study was to develop an high perf ormance liquid chromatography (HPLC) assay to validate the results of time-kill and PAFE experiments for voriconazole against C. albicans reference strains (ATCC 90029 and SC5314), and C. glabrata and C. parapsilosis bloodstream isolates (two each). Materials and Methods Antifungal Agents Stock solutions of voricon azole (Pfizer, New York, N. Y.) were prepared using sterile water. Stock solutions were separated into unit-of-use portions and stored at -80C until used. Test Isolates Eight strains were studied: 2 references (C. albicans: ATCC 90029 and SC 5314) and 6 clinical (3 C. albicans, 2 C. glabrata and 1 C. parapsilosis ).

PAGE 31

31 Antifungal Susceptibility Testing MICs were determ ined using E-test (Figur e 2-1). The inoculum was prepared from Sabouraud glucose agar subcultures inc ubated at 35C for 24 h and the resulting suspension was adjusted spectrophotometrically to a density equivalent to a 0.5 MacFarland standard at 530 nm (1.5 106 CFU/ml). The solidified medium was inoculated by dipping a sterile cotton swab into the respective undiluted stock inoculum suspension and streaking evenly in three directions over the en tire surface of a 150 mm diameter RPMI agar plate. The plate was pe rmitted to dry for at least 15 min before the E-test strips with antifungal were placed on the medium surface [46]. Antifungal Carryover Prior to tim e-kill experiments, assessment of the effect of solubilized voriconazole on colony count determinations. A fungal suspen sion was prepared with each test isolate to yield an inoculum of approximately 5 103 CFU/ml. One hundred-microliter volumes of these suspensions were added to 900l volum es of sterile water or sterile water plus voriconazole at concentrations ranging from 0.0625 to 16 times the MIC. This dilution resulted in a starting inocul um of approximately 5 102 CFU/ml. Immediately following addition of the fungal inoculum to a test tube the tube was vortexed and a 50-l sample was removed and plated without dilution on pot ato dextrose agar pl ates (Remel, Lenexa, Kans.) for determination of viable colony counts. Following 48 h of incubation at 35C, the number of CFU was determined. Tests we re conducted in quintuplicate. The mean colony count data for each agent at each multip le of the MIC tested were compared with the data for the control. Significant antifunga l carryover was defined as a reduction in the mean number of CFU per m illiliter of >25% compared w ith the colony count for the control [46].

PAGE 32

32 Time-kill Experiments Tim e-kill experiments were performed in duplicate at 0.25, 1, 4 and 16 MIC. Before testing, isolates were subcultured twic e on potato dextrose agar plates. Colonies from a 24to 48-h culture were suspended in 9 ml of sterile water and adjusted to a 0.5 McFarland turbidity standar d. 100 l of the adjusted fungal suspension was then added to either growth medium alone (control) or a solution of RPMI plus an appropriate amount of voriconazole stock solution. These procedures resu lted in a starting inoculum of approximately 1 105 to 5 105 CFU/ml concentrations and a voriconazole concentration of 0.0625, 0.25, 1, 4, or 16 MIC (Figure 2-2). Test solutions were placed on an orbital shaker and incubated with agitation at 35C. At predetermined time points of 2, 4, 8, 12, and 24 h, 100l samples were obtained from each solution, serially diluted in sterile water, and plated (100 l ) on potato dextrose agar plates for CFU determination. The lower limit of reproducib ly quantifiable CFU according to these methods was 50 CFU/ml [46, 119]. Postantifungal Effect (PAFE) Experiments PAFE experim ents for control testing (no drug) and testing at 0.25, 1, 4, and 16 times the MIC in duplicate tubes. After 1 hour of incubation, the cells were centrifuged at 1,400 g for 10 min, washed three times, and then resuspended in warm RPMI medium (9 ml) prior to reincubation. At predetermined time points of 2, 4, 8, 12, and 24 h, 100l samples were obtained from each solution, serial ly diluted in sterile water, and plated (100 l) on potato dextrose agar plates for CFU determination [119]. HPLC for Determination of Voriconazole in Stock Solution and RPMI Medium An HPLC protocol for measuring voricon azole in Stock Solution and RPMI m edia was developed based an existing assay me thod with a calibration range of 0.2 g/ml

PAGE 33

33 voriconazole in human plasma [120], 5 g/ml voriconazole in guinea pig plasma [121]. The analytical column was Kromasil C 18, 5 mm, 250.6 mm (Hichrom, Reading, UK) with a 10.2 mm guard cartridge (Hichr om, Reading, UK) packed with the same material at 25C in an Agilent 1100 Series apparatus. The mobile phase was acetonitrileammonium phosphate buffer (pH 6.0; 0.04 M) (1:1 v:v) and was degassed by filtration through a 0.45 mm nylon filter under vacuum. The flow rate was 0.8 ml/min and all chromatography was carried out at ambient temperature (~21C). Voriconazole concentrations were determined from peak areas detected by UV absorption at 255 nm with a retention time of 8.2 min; the maximum sensitivity was 0.025g/ml. Samples of Stock Solution and RPMI medium were diluted with 2 volum es of acetonitrileammonium phosphate buffer and centrifuged at full speed in a microcentrifuge for 10 min. The supernatants were applied to the column in 200l sample volumes. Calibration Curves for Voriconazole in Stock Solution and RPMI Medium Stock solutions of voriconazo le (1 g/l) were p repared in distilled water and diluted in RPMI medium to give 100 g/ ml. Standards were prepared by adding the diluted voriconazole solution to appropri ate volumes of Stock Solution and RPMI medium to give a concentrations of 0.025, 0.1, 0.2, 0.5, 1, 2, 4, 8 and 16 g/ ml. Calibration curves were constructed by plotti ng the peak area of voriconazole against concentration using a weighted (1:X2) least squares regression for HPLC data analysis. Statistical Analysis When data are expressed as the m ean SD, group mean differences were ascertained with analysis of variance (ANOV A). The results were considered significant if the probability of error was < 0.05.

PAGE 34

34 Results Stability of Voriconazole in Stock Solution and RPMI Medium The in vitro stability of voriconazole in stoc k solution (PH = 7.0) and RPMI m edium (PH= 7.0) were studied. There was no color change or precipitation in the preparations and pH remained stable during the period of the studies The results showed that voriconazole in stock solution were found stable at -80C, -4C at least for 6 months and one week, respectively. In RPMI medium voriconazole were found stable at room temperature and 37C for at least 72 hours, and stable -80C at least for 6 months (Table 2-1). The loss of voriconazole was less than 5% of the starting c oncentration at all conditions. Minimum Inhibitory Concentration (MIC) The MICs of all isolates were within the susceptible range, as measured by Etest and microdilution methods (Table 2-2) [122, 123]. Time-kills and Postantifungal Effect (PAFE) For tim e-kills and PAFEs, colonies from 48-hour cultures on Sabouraud dextrose agar (SDA) were suspended in 9 ml sterile water [46, 119]. One microliter of a 0.5 McFarland suspension was added to 10 ml of RPMI 1640 medium with or without voriconazole (0.25, 1, 4, and 16 MIC), an d the solution was incubated at 35C with agitation. The maximal voriconazole concentr ation in these experiments was 3.04g/ml (16 MIC for C. glabrata isolate 1). For time-kills, 100 l from each solution was serially diluted at desired time points ( 0, 2, 4, 8, 12, 24, 36, 48, 60, and 72 h) and plated on Sabouraud dextrose agar for colony e numeration. For PAFEs experiments, Candida cells were collected after 1 h of incubation, washed three times, and resuspended in warm RPMI 1640 medium (9 ml); colonies were enumerated at the desired time points.

PAGE 35

35 Voriconazole exhibited dose-re sponse effects against all Candida isolates during timekill experiments (Figure 2-3; Table 2-2), as higher concentrations resulted in greater growth inhibition or killing. The range of ma ximal growth inhibition of isolates at 1 and 4 MICs was -0.61to .78-log and .53to .99log, respectively, compared to those of controls. At 16 MIC, the range of maximal growth inhibition was .58to 4.15log (Table 2-3, 2-4, 2-5, 2-6, 2-7, 2-8 and 29). The results of PAFE experiments for the tested isolates are liste d in Tables of 2-10, 2-11, 2-12, 2-13, and 2-14. Voriconazole did not demonstrate PAFEs as shown from Figure 2-4 which is the profile of the concentration of colony forming units versus the time. Voriconazole at 16 MIC was fungicidal against C. parapsilosis isolate 2, reducing the starting inoculum by .21log at 24 h (fungicida l is defined as a > 2-log reduction of starting inoculum). Although kills did not ach ieve fungicidal levels for other isolates, voriconazole at 4 and 16 MIC reduced the starting inocula of C. glabrata isolate 2 and C. parapsilosis isolate 1 (Table 2-2). Of note, voricon azole was consistently fungistatic at 1 to 16 MICs, and the effect persisted for 72 h. Indeed, maximal inhibition of the four C. albicans and C. glabrata isolates at 4 and 16 MIC (c ompared to starting inocula) was not evident until 48 to 72 h. The two C. parapsilosis isolates, on the other hand, were maximally inhibited by 24 to 36 h. The time-kill curves of the C. parapsilosis isolates also differed from those of the other isolates at early time points. The C. parapsilosis isolates at 4 and 16 MIC were inhibited from entering the exponential growth phase, and dose-response effects were clearly evident by 8 h. The growth rates of C. albicans and C. glabrata isolates in the presence of voriconazole did not differ from those of controls during early exponential

PAGE 36

36 phase, but dose-response effects became incr easingly apparent as exponential growth continued (8 to 24 h). In our HPLC protocol for measuring voriconazole concentrations during time-kill and PAFE experiments, a 250by 4.6-mm an alytic column with 10by 3.2-mm guard cartridge (Hichrom, Reading, United Kingdom) was packed w ith a 5-m-particle-size Kromasil column at 25C in an Agilent 1100 series apparatus [120, 121]. Mobile phase acetonitrile-ammonium phosphate buffer (pH 6.0; 0.04 M; 1:1 vol:vol) was degassed by filtration through a 0.45-m nyl on filter under vacuum; the fl ow rate was 0.8 ml/min. Voriconazole concentrations were determin ed for peak areas detected by UV absorption at 255 nm with an 8.2-min retention time (Figure 2-5). For each isolate, we tested RPMI 1640 medium containing at least one dose of voriconazole between 1 and 16 MICs. Samples were diluted with 2 volumes of acetonitrile-ammonium phosphate buffer and centrifuged at full speed in a microcentrifuge for 10 min, and the supernatants (200 l) were applied to the column. The maximum sensitivity was 0.025g/ml, and the method produced linear results over a range of 0.025 to 12.8g/ml (r2 0.9996). In each instance, we confirmed that voriconazole concentrations remained constant throughout the duration of time-kill experiments (Figure 2-6), and the drug was fully removed during PAFE experiments (Figure 2-7). Discussion Voriconazo le were stable at all the cond itions covered in our studies. Our findings in time-kill studies conclusively demonstrate that voriconazole exerts prolonged fungistatic activity against C. albicans C. glabrata and C. parapsilosis at concentrations that are achievable in human sera with r outine dosing (the median s of the average and maximum voriconazole plasma concentrations in clinical trials were 2.51 and 3.79 g/ml,

PAGE 37

37 respectively) [124]. Our findings are potentially re levant clinically, since certain C. parapsilosis isolates exhibit diminished suscepti bility to echinocandin antifungals and C. glabrata isolates can develop resistance to fluc onazole and other antifungal agents. Although voriconazole caused >2-log kill of one C. parapsilosis isolate, further studies will be needed to accurately define the ex tent to which the drug might be fungicidal against clinical isolates. To our knowledge, this is the first study to verify standard time-kill and PAFE methodologies by directly measuring drug conc entrations. We describe a simple and reproducible HPLC method that has a broad, cl inically relevant dynamic range and does not require internal standards. The sensit ivity of voriconazole measurements within liquid media was greater than that previously reported for human or guinea pig plasma (0.2 to 10 and 5 to10 g/ml, respectively) [120, 121]. Based on our findings, we can assume that previous studies of azoles that showed fungistatic anticandida l activity and no PAFEs were conducted under the conditions of steady-state drug concentrations assumed by investigators. This demonstration is crucia l as efforts to use pharmacodynamic data to develop optimal antifungal treatment strategi es move forward. In particular, HPLC methods will be essential to the design of dynamic in vitro models to assess the pharmacodynamics of voriconazole and other ag ents prior to the achievement of steadystate conditions.

PAGE 38

38 Table 2-1. Percentage (%) of voriconazole maintenance in stock solution and RPMI medium at different temperatures (Mean SD, n = 3) Storage time1.02 mg/ml9.96 mg/ml0.20 g/ml1.02 g/ml10.13 g/ml 1mon (-80 C) 99.8 0.8 99.40.599.00.6100.10.9100.10.6 6mon (-80 C)98.41.798.80.697.51.797.41.099.50.6 1d (4 C)99.80.3 7d (4 C)98.60.7 24h (25 C) 99.20.799.00.6100.30.4 72h (25 C) 99.00.798.61.798.41.9 24h (37 C) 100.00.499.00.6100.30.6 72h (37 C) 97.00.997.41.697.70.7 VOR in stock solution VOR in RPMI medium Table 2-2. Minimum inhibitory concentration (MIC) of vor iconazole against Candida isolates (n=3) IsolateMIC(g/m l ) C. albicans ATCC90029 0.008 C.albicans SC5314 0.012 C. glabrata 1 0.19 C. glabrata 2 0.032 C. parapsilosis 1 0.008 C. parapsilosis 2 0.016

PAGE 39

39 Table 2-3. Voriconazole against Candida albicans ATCC90029 in constant concentration e xperiments (Data of CFU/mL, Mean SD, n = 3) Time (h)Control SD 0.25 MIC SD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.23E+055.13E+031.04E+056.20E+031.18E+053.76E+031.22E+051.21E+041.07E+052.21E+04 42.84E+059.81E+032.06E+051.28E+042.51E+053.50E+042.09E+053.44E+041.76E+055.22E+04 86.70E+059.24E+045.43E+051.14E+055.38E+054.33E+044.87E+055.72E+043.85E+054.89E+04 121.49E+062.71E+059.38E+051.52E+058.84E+054.75E+046.11E+059.06E+044.78E+052.23E+04 249.54E+062.57E+053.32E+061.19E+066.33E+051.81E+045.58E+057.55E+045.00E+056.81E+04 481.07E+075.78E+052.54E+062.00E+053.25E+051.17E+052.52E+058.03E+042.27E+051.14E+05 Table 2-4. Voriconazole against Candida albicans SC5314 in constant concentration experi ments (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD 0.25 MIC SD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 29.67E+041.71E+039.46E+041.51E+041.14E+058.02E+039.28E+041.15E+048.20E+049.21E+03 41.77E+052.11E+041.28E+056.96E+031.28E+052.66E+041.18E+051.97E+041.00E+052.11E+04 88.78E+052.44E+055.31E+051.92E+053.99E+051.35E+052.75E+058.31E+042.88E+051.10E+05 121.08E+061.96E+057.14E+054.46E+044.24E+051.70E+052.97E+053.88E+043.54E+051.53E+05 241.30E+061.06E+057.93E+055.74E+042.80E+056.57E+043.40E+051.40E+052.95E+051.13E+05 481.25E+061.55E+057.00E+059.44E+042.46E+059.56E+042.79E+058.14E+022.71E+056.26E+04

PAGE 40

40 Table 2-5. Voriconazole against Candida glabrata 1 in constant concentration experiments (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.06E+051.77E+041.03E+051.24E+041.08E+051.58E+041.05E+052.04E+041.00E+051.80E+04 41.15E+051.47E+041.11E+052.79E+041.46E+056.29E+041.19E+053.41E+041.31E+053.37E+04 81.76E+057.00E+041.54E+056.41E+041.63E+057.19E+041.61E+055.35E+041.70E+055.59E+04 125.25E+056.45E+035.35E+054.15E+044.22E+051.07E+052.85E+052.55E+043.61E+052.85E+04 242.46E+065.70E+041.39E+062.76E+057.32E+051.80E+056.96E+051.52E+056.20E+051.10E+04 483.86E+062.49E+052.19E+063.90E+041.05E+064.87E+057.45E+058.02E+044.36E+054.11E+04 Table 2-6. Voriconazole against Candida glabrata 2 in constant concentration experiments (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 29.60E+043.97E+021.01E+051.63E+041.01E+053.62E+038.67E+041.41E+049.96E+041.83E+04 41.20E+052.64E+031.25E+052.84E+041.24E+055.68E+031.11E+051.26E+041.11E+052.68E+04 84.86E+052.51E+054.28E+051.66E+053.52E+057.74E+042.78E+058.54E+042.81E+059.87E+04 121.23E+064.34E+058.90E+056.26E+046.03E+051.25E+053.37E+054.88E+042.94E+054.36E+04 245.34E+069.61E+051.08E+062.01E+055.80E+052.78E+044.26E+051.39E+043.52E+056.21E+04 487.11E+061.41E+062.96E+051.42E+056.91E+043.16E+043.74E+041.26E+043.47E+041.44E+04

PAGE 41

41 Table 2-7. Voriconazole against Candida parapsilosis 1 in constant concentration experiment s (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.06E+058.26E+039.41E+042.25E+049.62E+048.26E+031.00E+056.56E+031.03E+052.44E+03 41.67E+056.03E+031.41E+051.33E+041.39E+052.05E+021.36E+053.57E+031.30E+058.87E+03 89.97E+059.36E+036.39E+051.98E+042.71E+055.42E+042.30E+053.12E+031.94E+053.28E+03 121.84E+064.52E+051.64E+063.35E+053.74E+052.73E+041.48E+057.51E+041.25E+054.19E+04 247.26E+062.44E+065.32E+061.51E+066.67E+051.41E+051.25E+058.30E+031.21E+051.34E+04 483.45E+072.51E+062.72E+071.06E+073.47E+065.73E+051.28E+053.71E+049.06E+043.60E+04 Table 2-8. Voriconazole against Candida parapsilosis 2 in constant concentration experiment s (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.10E+057.07E+031.04E+051.41E+039.49E+043.18E+039.55E+042.33E+048.55E+047.00E+03 41.36E+051.70E+041.25E+052.12E+031.09E+051.27E+041.08E+059.90E+038.00E+042.68E+04 83.51E+051.77E+044.25E+057.85E+047.44E+041.58E+046.47E+041.88E+042.32E+047.79E+03 121.16E+062.05E+059.15E+059.69E+048.09E+042.00E+043.07E+049.76E+031.26E+041.41E+03 245.60E+063.04E+056.08E+062.57E+067.12E+043.04E+031.15E+042.98E+032.78E+033.07E+03 481.02E+072.21E+061.42E+074.53E+065.35E+055.69E+055.15E+042.87E+041.73E+037.14E+02

PAGE 42

42 Table 2-9. Summarized voriconazole time-kill data against Candida isolates Maximum log kill ata: IsolateMIC(g/ml)b24h 24-48h >48h 1 x MIC4 x MIC16 x MIC1 x MIC4 x MIC16 x MIC1 x MIC4 x MIC16 x MIC C. albicans ATCC900290.008-1.23-1.34-1.34 -1.77 -1.84 -1.96 -1.63 -1.90 -1.89 C.albicans SC5314 0.012 -0.61 -0.51-0.54-0.56 -0.53-0.58 C. glabrata 1 0.19-0.43-0.49-0.59-0.74-0.78-1.00 -0.99-1.08-1.17 C. glabrata 2 0.032-1.02-1.10-1.15-2.23-2.39-2.49 -2.78-2.99-3.02 C. parapsilosis 10 0 0 8 -0.99 -1.64-1.68-0.91 -2.67-2.74 -0.67-2.37-2.69 C. parapsilosis 2 0.016 -1.86-2.70 -3.93-1.03-2.13-3.94-0.88-1.12 -4.15 a Time-kill data are presented as the maximum difference between the growth of the cont rol isolate (grown in the absence of voric onazole) and that of the isolate in the presence of various voriconazole concentrations at 24 h, 24 to 48 h, and > 48 h. The maximum growth inhibition of each isolate at the given concentrations is shown in bold. b MICs, as determined by Etest and the Clinical and Laboratory Standards Institute (formerly NCCLS) broth microdilution method [123], were within twofold agreement. Table 2-10. Voriconazole against Candida albicans ATCC90029 in PAFE experiments (Data of CFU/mL, Mean SD, n =2) Time (h)Control SD0.25 MICSD1 MICSD4 MICSD16 MICSD 03.00E+051.98E+043.04E+051.56E+042.95E+057.78E+033.15E+057.07E+032.87E+051.48E+04 23.20E+052.83E+033.15E+052.05E+042.89E+052.12E+033.28E+055.66E+033.35E+056.36E+04 45.65E+057.07E+036.15E+051.48E+055.15E+053.54E+046.20E+051.27E+055.75E+051.34E+05 81.75E+061.41E+051.99E+067.64E+051.66E+062.47E+051.91E+063.11E+052.17E+061.20E+06 125.10E+062.83E+055.40E+064.24E+054.65E+067.07E+044.35E+062.12E+055.30E+067.07E+05 241.55E+071.27E+061.80E+073.89E+061.46E+071.06E+061.60E+074.17E+061.61E+073.68E+06 482.56E+072.69E+062.84E+077.07E+062.77E+074.88E+062.24E+071.34E+062.93E+075.02E+06

PAGE 43

43 Table 2-11. Voriconazole against Candida glabrata 2 in PAFE experiments (Data of CFU/mL, Mean SD, n =2) Time (h)Control SD0.25 MICSD1 MICSD4 MICSD16 MICSD 02.61E+051.77E+042.31E+052.12E+032.38E+054.24E+032.62E+054.95E+032.44E+051.27E+04 22.48E+052.47E+042.18E+054.81E+042.66E+051.48E+042.43E+055.23E+042.30E+051.77E+04 45.55E+052.12E+044.80E+054.24E+045.20E+054.24E+046.00E+052.83E+045.30E+052.83E+04 81.37E+067.07E+041.48E+061.34E+051.20E+067.07E+031.58E+069.90E+041.79E+061.20E+05 123.19E+066.36E+043.08E+068.49E+043.35E+062.12E+053.00E+061.41E+053.60E+061.06E+05 241.48E+077.07E+051.59E+072.76E+061.80E+074.81E+061.54E+076.36E+051.67E+074.24E+05 481.91E+075.66E+051.54E+071.41E+051.75E+071.34E+061.72E+073.54E+051.97E+071.56E+06 Table 2-12. Voriconazole against Candida glabrata 1 in PAFE experiments (Data of CFU/mL) Time (h)Control 0.25 MIC1 MIC4 MIC16 MIC 02.58E+052.35E+052.58E+052.67E+052.77E+05 22.37E+052.15E+052.37E+052.39E+052.64E+05 43.40E+053.07E+053.53E+052.60E+053.10E+05 88.60E+056.70E+051.06E+067.00E+051.03E+06 121.41E+061.62E+061.53E+061.05E+061.15E+06 247.80E+067.00E+067.00E+067.80E+066.80E+06 481.15E+071.29E+071.35E+071.16E+071.34E+07

PAGE 44

44 Table 2-13. Voriconazole against Candida parapsilosis 1 in PAFE experiments (Data of CFU/mL) Time (h)Control 0.25 MIC1 MIC4 MIC16 MIC 02.89E+053.40E+053.10E+052.70E+052.84E+05 22.51E+052.85E+052.95E+052.46E+052.67E+05 44.80E+056.80E+056.40E+056.00E+056.27E+05 89.90E+051.69E+061.12E+061.55E+061.70E+06 122.40E+063.42E+062.50E+063.04E+063.16E+06 241.46E+072.18E+071.47E+071.64E+072.00E+07 483.22E+073.04E+073.02E+073.03E+073.69E+07 Table 2-14. Voriconazole against Candida parapsilosis 2 in PAFE experiments (Data of CFU/mL) Time (h)Control 0.25 MIC1 MIC4 MIC16 MIC 01.81E+051.79E+051.42E+051.55E+051.79E+05 21.38E+051.36E+051.40E+051.46E+051.34E+05 41.34E+051.61E+051.54E+052.20E+051.40E+05 84.10E+056.00E+053.66E+056.50E+054.70E+05 121.34E+061.40E+061.33E+061.92E+061.48E+06 241.37E+071.60E+071.45E+071.95E+071.21E+07 482.26E+072.29E+072.36E+072.43E+072.00E+07

PAGE 45

45 A B Figure 2-1. Susceptibility of vor iconazole against Candida isolates were determined by disc diffusion, E-test and the Clinical and Laboratory Standards Institute broth microdilution method. A) Di sc diffusion. B) E-test. Figure 2-2. Culture flasks for voric onazole against Candida isolates in vitro

PAGE 46

46 C. albicans ATCC90029 Time_hr 04812162024283236404448 N_CFU/mL 1e+4 1e+5 1e+6 1e+7 1e+8 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. albicans sc5314 Time_hr 04812162024283236404448 N_CFU/mL 1e+4 1e+5 1e+6 1e+7 1e+8 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. glabrata 1 Time_hr 04812162024283236404448 N_CFU/mL 1e+4 1e+5 1e+6 1e+7 1e+8 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. glabrata 2 Time_hr 04812162024283236404448525660646872 N_CFU/mL 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. parapsilosis 1 Time_hr 04812162024283236404448525660646872 N_CFU/mL 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. parapsilosis 2Time_hr 04812162024283236404448 N_CFU/mL 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC Figure 2-3. Time-kill curves for voriconazole against Candida isolates (Mean SD, n=3 for C. albicans ATCC90029, n=2 for other tested strains, without significant differences in results by ANOVA)

PAGE 47

47 C. glabrata 1Time_hr 04812162024283236404448 N_CFU/ml 1e+5 1e+6 1e+7 1e+8 ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. glabrata 2Time_hr 04812162024283236404448 N_CFU/mL 1e+5 1e+6 1e+7 1e+8 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. parapsilosis 1Time_hr 04812162024283236404448 N_CFU/ml 1e+5 1e+6 1e+7 1e+8 ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. parapsilosis 2Time_hr 04812162024283236404448 N_CFU/ml 1e+5 1e+6 1e+7 1e+8 ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC C. albicans ATCC90029Time_hr 04812162024283236404448 N_CFU/mL 1e+5 1e+6 1e+7 1e+8 Ctrl 0.25*MIC 1*MIC 4*MIC 16*MIC Figure 2-4. Voriconazole did not demonstrate PAFEs. (Mean SD, n=2 for C. albicans ATCC90029, C. glabrata 2, without si gnificant differences in results by ANOVA; n=1 for other three tested strains.)

PAGE 48

48 min 0 2 4 6 8 10 12 mAU 0 25 50 75 100 125 150 175 DAD1 A, Sig=255,4 Ref=360,100 (IVY5J07\011-1101.D ) Area: 2 0 09.46 8.141 Figure 2-5. Chromatogram of voriconazole which concentration was determined from peak area detected by UV absorptio n at 255 nm with around an 8.2 minute retention time Time_hr 04812162024283236404448 % Voriconazole Concentration 0 20 40 60 80 100 120 Figure 2-6. Voriconazole concentrations in culture media thr oughout the duration of time-kill experiments by HPLC: Represen tative data for a 48-h time-kill experiment with C. glabrata isolate 1 are presented. (Mean SD, n=3, without significant differe nces in results by ANOVA)

PAGE 49

49 Time_hr 04812162024283236404448 % Voriconazole Concentration -50 -40 -30 -20 -10 0 10 20 30 40 50 Figure 2-7. Voriconazole concentrations in PAFE experiments by HPLC: Representative data for a 48-h time-kill experiment with C. glabrata isolate 1 are presented. (Mean SD, n=3, without significant differences in results by ANOVA)

PAGE 50

50 CHAPTER 3 APPLYING PHARMACOKINETIC/PHARM ACODYNAMIC MATHEMATICAL MODEL ACCURATELY DESCRIBES TH E ACTIVITY OF VORICONAZOLE AGAINST CANDIDA SPP. IN VITRO Background Tim ekill curves are attractive tools for studying the pharmacodynamics of antimicrobial agents as they provide detailed information about antimicrobial efficacy as a function of both time and concentration [125]. Although timekill curves can be studied using animal models of infection, in vitro models offer significant advantages in cost, convenience and time, as well as permitting di rect investigation of the antimicrobial microbe interaction in a cont rolled and reproducible manner [126]. The results of specific timekill experiments can be accurately described using pharmacokinetic/ pharmacodynamic (PK/PD) mathematical mode ls. By using PK data derived from clinical trials or animal experiments, mathem atical models can be used to simulate the expected kill curves for different doses a nd dosing regimens of an antimicrobial agent [127]. As such, PK/PD modelling is a potential ly powerful technique for defining optimal antimicrobial treatment strategies. Voriconazole is a triazole antifungal agen t effective in the treatment of patients with candidiasis [8, 65]. We recently performed timekill experiments against two strains each of Candida albicans Candida parapsilosis and Candida glabrata using voriconazole concentrations of 0.25, 1, 4 and 16 the minimum inhibitory concentration (MIC) and measuring col ony counts at 0, 2, 4, 8, 12, 24, 36, 48, 60 and 72 h [22]. We demonstrated that voriconazole ex hibited doseresponse effects against all isolates, as higher concentrations resulted in enhanced killing of Candida spp. The ranges of maximal growth inhibition of isolates at concentrations of 1 and 4 MIC were 0.61

PAGE 51

51 to 2.78 log and 0.53 to 2.99 log, respectively, compared with controls. At 16 MIC, the range of maximal growth inhibition was 0.58 to 4.15 log. Specific Aim The objectives of the present study w ere to develop a PK/PD mathematical model to fit our timekill data and to use the mode l to simulate the expected kill curves for typical intravenous and oral dosing regimens. Materials and Methods Mathematical Modelling of Timekill Data The m ethods and results of antifungal susceptibility testing and timekill curve experiments for voriconazole against Candida isolates were described previously [22]. During timekill experiments, aliquots were ta ken at each time point and examined under the microscope to exclude hyphal formation or clumping. The voriconazole MICs against the isolates ranged from 0.008 to 0.19g/mL. Fitting of timekill data was performed with a series of models us ing the Scientist 3.0 non-lin ear least squares regression software program (MicroMath, Salt Lake City, UT), as detailed in the Results and Discussion section. To determine the model that best fits the data for each isolate, graphs were visually inspected for goodness of fit and several criteria were considered, including model selection criterion (MSC), coefficient of determination ( R2) and the correlation between measured and calculated data points. Simulations of Expected Timeki ll Curves Using Human PK Data PK param eters were extracted from published data sets of humans receiving voriconazole using XyExtract v2.5(Wilton and Cl eide Pereira da Silva, Campina Grande, Para ba, Brazil) and the data were analyzed by WinNonlin 5.2 (Pharsight Corporation, Mountain View, CA) to describe the best absorption model. The Scientist 3.0 software

PAGE 52

52 program was then used to simulate plas ma concentrationtime profiles for multiple dosing of voriconazole ( = 12 h). Simulations of expected kill curves for each isolate were made using steady-state PK parameters in the mathematical model that was shown earlier to best fit the timekill data. Statistical Analysis When data are expressed as the m ean SD, group mean differences were ascertained with analysis of variance (ANOV A). The results were considered significant if the probability of error was < 0.05. Results An Adapted Sigmoidal Emax Model Provides the Best Fit for Voriconazole Timekill Data against Candida Isolates Fitting of the timekill data was started with a simple, commonly used Emax model [127]: N CEC Ck k dt dNs 50 max (3-1) In this model, dN /d t is the change in number of Candida as a function of time, ks (h 1) is the candidal growth rate constant in the absence of voriconazole, kmax (h 1) is the maximum killing rate constant (i.e. maximum effect),EC50 (mg/L) is the concentration of voriconazole necessary to pr oduce 50% of maximum effect, C (mg/L) is the concentration of voriconazole at any time ( t ) and N (colony-forming units (CFU)/mL) is the number of viable Candida. Since voriconazole concentrat ions did not change during the timekill experiments, C was constant for the en tire fitted time period. The Emax model does not account for the fact that isolates have not yet reached the logarithmic growth phase at time zero or for the delay in the onset of killing by

PAGE 53

53 voriconazole. To compensate for these de lays, an exponential correction factor (1 exp zt) was incorporated into the preceding model. Since the delays in the onset of candidal growth and voriconazole killing were not n ecessarily equivalent, they were assigned different adaptation rate consta nts (i.e. delay in growth = (1 exp t) and delay in killing = (1 exp t)): Ne CEC Ck ek dt dNt t s 1 150 max (3-2) Model (2) was adapted by considering the maximum number of Candida ( Nmax), which potentially corrected for the limitations of space and nutrients that are inherent in in vitro systems: Ne CEC Ck e N N k dt dNt t s 1 1 150 max max (3-3) In the final model, a Hill factor ( h ) was incorporated, which modified the steepness of the slopes and smoothed the curves: Ne CEC Ck e N N k dt dNt h h h t s 1 1 150 max max (3-4) The fitted timekill curves of the six Candida isolates were generated using the different models in the absence of voriconazole (control) or in the presence of constant concentrations (mg/L) of voriconazole. In Model (4) Hill factor values ranging from 0 to 4 were tested. In general, the curve fittings for all PD data were acceptable. The best fits for each isolate were obtained using Model (4); the PD parameters and goodness of fit criteria were excellent for all voriconazole Candida isolate pairings, with MSC 2.46 and R2 0.95. The fitted curves for Model (4) are shown in Figure 3-2. The PD parameters of

PAGE 54

54 voriconazole against the six Candida isolates and the goodness of fit criteria are presented in Table 3-1 In timekill experiments, PD effects can be described by at least three parameters: k s, the growth rate constant of th e organism in the absence of drug; k max, the maximum killing rate constant of th e drug against the organism; and EC50 (the concentration of drug necessary for 50% of its maximum effect) [125, 127]. The EC50 values calculated in the present study demonstrate th at voriconazole was highly effective at easily achievable serum levels against each of the six Candida isolates (EC50 range 0.002.05 mg/L) ( Table 3-1 ). Moreover, the rates of maximal killing by voriconazole were highly correlated with the growth rates of the is olates (Pearsons correlation coefficient = 0.9861). In other words, the data suggest that voriconazole achieved its highest kill rates against the most rapidly proliferating organisms. Human PK Data for Voriconazole Can Be Used to Simulate Expected Timekill Curves Mathematical models can be us ed to simulate the expected kill curves for different doses and dosing regimens of antimicrobials by substituting the concentration term with PK parameters collected in vivo and accounting for protein binding [127]. To simulate timekill curves that mi ght be expected for voriconazole under typical dosing regimens, we first extracted PK parameters from published data sets of humans receiving the drug intr avenously (3 mg/kg twice a da y (bid)) and orally (200 mg bid) in steady-state [39]. Plasma concentration profiles for intravenous and oral dosing were found to be best charac terized by two-compartment mode ls of zeroand first-order absorption, respectively. Some of the PK parameters are listed in Table 3-2

PAGE 55

55 Applying the PK parameters for the two dosing regimens, we simulated plasma concentrationtime profiles for multiple dosing of voriconazole ( = 12 h), assuming plasma protein binding of 58% ( Figure 3-2 ) [8]. We then used the steady-state PK parameters to generate simulations for voriconazole activity against each of the Candida isolates in the present study ( Figure 3-2 ). As shown, voriconazole showed fungistatic activity against all si x strains. For five of the six strains, simulated kills were virtually identical for the intravenous and oral regimens. The exception was C. parapsilosis 2, for which the effects of intravenou s dosing were greater By collecting PK data for a range of voriconazole doses and dosing regimens during clinical trials or animal models, we should be able to use the mathematical model to predict treatment strategies that might be most effective against a given Candida isolate. Discussion A pharmacokinetic/pharmac odynamic (PK/PD) mathematical model was also developed in our studies to fit voriconazole timekill data against Candida isolates in vitro and used the model to simulate the expect ed kill curves for typical intravenous and oral dosing regimens. Although PD studies comparing the efficacy of various voriconazole regimens against Candida isolates are feasible in animal models [32], they are laborious, expensive and complicated. In vitro models permit direct study of the interaction between fungi and fungal agents in a controlled and reproducible manner and allow direct comparisons of different agents and dosing strategies in a more convenient and faster way [126]. We have shown that it is feasible to describe precisely the timekill activity of voriconazole against Candida isolates by using an adapted sigmoidal Emax model that corrects for delays in candidal growth and the onset of voricon azole activity, saturation of

PAGE 56

56 the number of Candida that can grow in the in vitro system and the steepness of the concentrationresponse curve. We establis hed the PK/PD models and simulated the expected voriconazole activity for typical intravenous and oral dosing regimens by using PK data collected in vivo and accounting for protein bindi ng. A limitation of the present study design was that voriconazo le concentrations were constant. In the future, we will characterize voriconazole Candida interactions using a dynamic in vitro system in which drug concentrations change ove r time in a manner consistent with the PK profile in humans. Based on the success of the present st udy, it should also be feasible to model accurately the timekill data for changing voric onazole concentrations. In conclusion, we demonstrated that the activ ity of voriconazole against Candida isolates can be accurately described using a mathematical model. We anticipate that this approach will be an efficient method for devising optimal voric onazole treatment strategies against candidiasis.

PAGE 57

57 Table 3-1. Pharmacodynamic parameters and goodness of fit criteria against Candida isolates Parameter(unit) C. abicans ATCC90029 C. abicans SC5314 C. g labrata 1 C. g labrata 2 C. p ara p silosis 1 C. p ara p silosis 2 C (mg/L) 0.002/0.008/0.032/0.1280.003/0.012/0.048/0.1920.0475/0.19/0.76/3.040.008/0.032/0.128/0.5120.002/0.008/0.032/0.1280.004/0.016/0. 064/0.256 Ks (h-1) 0.28 0.87 0.24 0.26 0.22 0.66 Kmax (h-1) 0.34 0.72 0.29 0.35 0.24 0.63 EC50 (mg/L) 0.00150.00300.050.00420.00660.012 alpha1.240.190.140.580.260.072 beta 0.069 0.11 0.032 0.059 0.23 0.19 Nmax (CFU/mL) 1.09E+07 1.21E+06 3.85E+06 6.57E+06 3.48E+07 6.95E+06 h 2.13 2.77 1.16 2.32 1.49 3.84 MSC/R^2 4.68/0.99 3.58/0.99 4.76/0.995 4.33/0.99 5.63/0.997 4.91/0.95 ks: fungal growth rate constant in the absence of voriconazole kmax: maximum killing rate constant (maximum effect) EC50: concentration of voriconazole necessa ry to produce 50% of maximum effect alpha: constant used to fit the in itial lag phase for the growth beta: constant used to fit the initial lag phase for the inhibition or killing h: Hill factor MSC: model selection criteria

PAGE 58

58 Table 3-2. Steady-state pharmacokinetic para meters in plasma as calculated by twocompartment model analysis Dose regimenParameter (unit)200mg oral 3mg/kg/h i.v Cmax (mg/L) 1.592.66 Tmax (h) 1.721.00 (h-1) 0.4431.260 (h-1) 0.1390.120 F0 91 0 CL/F (L/h)17.914.2 Vdss/F (L) 98.9109.7 Vdarea/F (L) 128.5117.9 i.v. : intravenous Cmax: maximum concentration Tmax: time to Cmax : macro rate constant associat ed with the distribution phase : macro rate constant associated with the elimination phase F : bioavailability CL/ F : systemic clearance Vdss/ F : apparent volume of di stribution at steady-state Vdarea/ F : apparent volume of distribution

PAGE 59

59 2 compartment i.v. short-term infusion 3mg/kg/hTime post-dose (h) 0102030405060 Mean plasma concentration (ng/ml) 0 1000 2000 3000 4000 5000 VORI_total VORI_free 2 compartment 200mg oralTime post-dose (h) 0102030405060 Mean plasma concentration (ng/ml) 0 500 1000 1500 2000 2500 3000 VORI_total VORI_free Figure 3-1. Plasma VOR concen tration-time profiles simulated with a two-compartment PK model. (PK data were extracted from published paper[39]). Short-term intravenous (3mg/kg/h, bid, T=1h); Oral administration (200mg, bid, F=0.9); fb = 58%

PAGE 60

60 C. albicans ATCC90029Time_hr 01 22 43 64 8 N_CFU/mL 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 1 01 22 43 64 8 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.0475mg/L N3:0.19mg/L N4:0.76mg/L N5:3.04mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 2 0122436486072 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.008mg/L N3:0.032mg/L N4:0.128mg/L N5:0.512mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 1 01 22 43 64 8 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 2 01 22 43 64 8 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 N1(control) N2(0.004mg/L) N3(0.016mg/L) N4(0.064mg/L) N5(0.256mg/L) N1_cal N2_cal N3_cal N4_cal N5_cal C. albicans SC5314 01 22 43 64 8 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L N1_cal N2_cal N3_cal N4_cal N5_cal Figure 3-2. Fitted time-kill curves derived us ing the mathematical model for constant concentrations of voriconazo le (Mean SD; n = 3 for Candida albicans ATCC90029 and n = 2 for the other strains, without significant differences in results by analysis of variance (ANOVA)). Voriconazole concentrations were 0, 0.25, 1, 4 and 16 the minimum i nhibitory concentrations (listed as N1-N5, respectively, in th e individual figure legends).

PAGE 61

61 C. albicans ATCC90029Time_hr 0102030405060 N_CFU/mL 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8C_mg/L 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 ctrl growth oral kill IV kill C total _oral C free _oral C total _iv C free _iv C. albicans SC5314 0102030405060 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 Figure 3-3. Simulations of candidal timek ills and plasma voriconazole concentration time profiles. The simulations of timekills are shown for short-term intravenous ( i.v.) infusion () and oral dosing of voriconazole (---). Control growth is in the absen ce of drug (). The simulated plasma concentration time profiles for multiple doses of voriconazole ( = 12 h) are shown for short-term i.v. infusion ( = 1 h, infusion rate = 3 mg/kg/h, twice a day (bid)) and oral administration (200 mg, bid, F = 0.9). Free voriconazole concentrations were calculated assumi ng plasma protein binding level of 58%. C. glabrata 1 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. glabrata 2 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. parapsilosis 1 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. parapsilosis 2 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. glabrata 1 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. glabrata 2 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. parapsilosis 1 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. parapsilosis 2 01 02 03 04 05 06 0 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e-2 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8

PAGE 62

62 CHAPTER 4 PHARMACODYNAMIC STUDY IN DYNAMIC SYSTEM FOR DESCRIBING THE ACTI VITY OF VORICONAZOLE AGAINST CANDIDA SPP. IN VITRO Background Voriconazole is a lipophilic triazole antifungal agent that exhibits potent activity in vitro and in vivo against a variety of fungi including Candida spp., Aspergillus spp., Cryptococcus neoformans Blastomyces dermatitidis, Coccidioides immitis Histoplasma capsulatum Fusarium spp. and Penicillium marneffei [41-43]. Using standard in vitro time-kill methodologies, we and others have shown that voriconazole ex erts prolonged fungista tic activity and no postantifungal effect against diverse Candida spp., including C. albicans, C. glabrata and C. parapsilosis [8, 22]. In a follow-up study, we developed a mathematical model that accurately described the activity of voriconazole against Candida spp. during time-kill experiments [102]. After identifying and validating the in vitro pharmacodynamic (PD) parameters derived from our experiments, we used pharmacokine tic (PK) data extracted from human data sets to simulate the expected kill curves in vivo for typical intravenous and oral dosing regimens of voriconazole. Indeed, by using PK data from huma n clinical trials or animal studi es to simulate kill curves for different doses and dosing regimens of voriconazole or other antimicrobial agents, PK/PD modeling is potentially a power ful technique for defining opt imal treatment strategies [127]. The major limitation of our earlier studies was that voriconazole concentrations remained constant over time at fixed multiples of the minimum inhibitory concentration (MIC) [22, 46, 102, 128, 129]. Specific Aim The first objective of the pres ent study was to develop a dynamic time-kill methodology in which voriconazole concentrations diminished in a manner consistent with the serum PK profile in humans. After we characterized the in teraction between vori conazole and five Candida

PAGE 63

63 isolates using the dynamic in vitro system, our objectives were to fit our mathematical model to the changing concentration kill-curves and to use hum an PK data to simulate expected kill curves in vivo Materials and Methods Antifungal Agents Stock solutions of voriconazo le (Pfizer, New York, N.Y.) were prepared using sterile water. Stock solutions were separated into unit-of-use portions and stored at -80C until used. Test Isolates Five strains were studied: 1 reference ( Candida albicans ATCC 90029) and 4 clinical (2 Candida glabrata and 2 Candida parapsilosis). Softwares PK parameters were extracted from publishe d data sets of humans receiving voriconazole using XyExtract v2.5(Wilton a nd Cleide Pereira da Silva, Campina Grande, Para ba, Brazil) and the data were analyzed by WinNonlin 5.2 (P harsight Corporation, Mountain View, CA) to describe the best absorption model. The Scientist 3.0 software program was then used to simulate plasma concentrationtime profiles for multiple dosing of voriconazole ( = 12 h). Dynamic Model Design In the infection model, the voriconazole was introduced as an intravenous bolus and was eliminated at constant rate according to the first order kinetics: tkeeCC0 (4-1) 0V A C (4-2)

PAGE 64

64 Where C0 is the initial voriconazole concentration in the flask, t is the time that has elapsed since the loading of voriconazole, C is th e voriconazole concentration at time t and ke is the voriconazole elimination rate constant. For any two points of time t 101 tkeeCC (4-3) 20 2 tkeeCC (4-4) 0 1 1V A C (4-5) 0 1 0 2 21 V VCA V A C (4-6) From equations 4-3, 4-4, 4-5 and 46, equation 4-7 could be derived: 0 0 0 01 2 1V VeC eCeCtk tk tke e e (4-7) Reconstitute equation 4-7, )1()( 021ttkeeVV (4-8) As t = t1t2, )1(0 tkeeVV (4-9) Where V0 is the initial volume of medium in the flask, and t is the time period of replacing voriconazole-free medium. The Ke could be calculated usi ng a certain half-life (t1/2): 2/12ln t Ke (4-10)

PAGE 65

65 As shown in Figure 4-1, every 2 hours ( t), 4.13 ml ( V) of voriconazo le containing medium were replaced by the same volume of vori conazole-free medium in order to simulate an in vivo half-life of 6 hours. Time-kill Experiments in the Dynamic Model The MICs of voriconazole against a C. albicans reference strain (ATCC 90029) and C. glabrata and C. parapsilosis clinical isolates (two each), as determined by E-test and the Clinical and Laboratory Standards Institute (formerly NCCLS) broth microdilu tion method, were consistent with previous reports [22, 46, 123]; values were within two-fold agreement by the two methods (Table 2-1). In our previous timekill experiments, voriconazole at constant concentrations of 1, 4 and 16MIC inhibited growth of the isolates in a dose-dependent manner [22]. The dynamic time-kill experiments were pe rformed in duplicate against each isolate (inocula of 15 to 505 CFU/mL). Colonies from 24-hour cultures on Sabouraud dextrose agar (SDA) plates were suspended in 9 mL of sterile water a nd adjusted to UV absorbance of 0.84-0.86 at 535nm; 200L of the su spension were added to 20 mL (V0) of RPMI 1640 (Sigma Chemical Co.) buffered to a pH of 7.0 w ith MOPS (morpholinepropanesulfonic acid; RPMI 1640 medium) without voriconazole, and incubated at 35C with continuous agitation. After 2 hours pre-incubation, voriconazole was added at starting concentr ations of 0.25, 1, 4, or 16MIC. The drug was introduced as a bolus and e liminated at constant rate according to first order kinetics which was descri bed previously. Prior to the re moval of voriconazole-containing medium at each time point, 100 L aliquots were obtained from each solution, serially diluted, and plated on SDA plates to enumerate CFUs. Voriconazole concentrations during time-kill experiments were verified by serial HPLC measurements [22].

PAGE 66

66 Results Voriconazole exhibited dose-re sponse effects against all Candida isolates during the dynamic time-kill experiments, as higher starting concentrations resulted in greater growth inhibition or killing (Figures 4-2). The time-kill data in the dynamic time-kill experiments for individual isolates were listed in the tables of Table 4-1, 4-2, 4-3, 4-4, 4-5. Of note, starting concentrations of 4 and 16MIC resulted in growth inhibition or killing of all five isolates; results with 16MIC starting c oncentrations were clearly superior to 4MIC starting concentrations for four isolates ( C. albicans ATCC90029, C. glabrata 2, C. parapsilosis 1, and C. parapsilosis 2). On balance, the results were consistent with those observed for constant concentrations of 4 and 16MIC against the same isolates in our earlier time-kill study [22]. In the present study, results with 1 MIC starting concentrations were significantly inferior to 4 and 16MIC against all five isolat es (Figure 4-2). In fact, 1 MIC starting concentrations caused slight growth inhibiti on of only two strains ( C. albicans ATCC90029 and C. glabrata 1); for the remaining three isolates, the 1 MIC growth curv es did not differ significantly from controls grown in the absence of voriconazole. These results differed from constant concentrations of 1 MIC in the previous study, for which growth inhi bition and kills of all isolates were significantly compared to controls. In fact, growth inhibition and kills at constant concentrations of 1 MIC approached those seen at constant concentratio ns of 4 and 16MIC against three isolates ( C. albicans ATCC90029, C. glabrata 1 and C. glabrata 2). We used our previously developed sigmoidal Emax-model to fit time-kill data for each voriconazole-Candida pairing in the dynamic experiments [102, 127]. The model accounts for delays in the onset of both Candida growth a nd voriconazole-induced ki ll, the saturation of Candida growth, and incorporates a Hill factor th at modifies the steepness of the slopes and smoothes the curves:

PAGE 67

67 Ne CEC Ck e N N k dt dNt h h h t s 1 1 150 max max (4-11) In this model, d N /d t is the change in number of Candida as a function of time, ks (h 1) is the candidal growth rate constant in the absence of voriconazole; kmax (h 1) is the maximum killing rate constant; EC50 (mg/L) is the concentration of vor iconazole necessary to produce 50% of maximum effect; C (mg/L) is the concentration of voriconazole at any time ( t ); N (CFU/mL) is the number of viable Candida; is the constant used to fit th e initial lag phase for the growth; is the constant used to fit the initial lag phase for the inhibition or killing; and h is the Hill factor. Simultaneous fitting of the model to the changing concentration kill-curves was performed using Scientist3.0 non-linear l east squares regression software program (MicroMath, Salt Lake City, UT) as previously described [102]. Graphs were visually inspected for quality of fit and several criteria were considered including model selection criterion (MSC), coefficient of determination (R2) and the correlation between measured and calculated data points. For each isolate, goodness of fit crite ria was excellent, with MSC 2.71 and R2 0.97 (Table 4-6, Figure 4-3). Similarly, PD parameters we re excellent (Table 4-6). The EC50 values (drug concentrations necessary for 50% of maximum effect) indicate that voriconazole was highly effective against each of the five isolates at easily achievable serum levels (range: 0.005 to 0.08 mg/L); these values were comparable to those obtained in the earlier constant concentration experiments (range: 0.0015 to 0.05 mg/L). As in the previous study, vor iconazole achieved its highest rates of kill against the most rapidly proliferating isolates. To simulate the expected kill curves for different doses and dosing regimens of voriconazole, we substituted the co ncentration term in our model with PK parameters collected

PAGE 68

68 in vivo [39, 102]. PK data were extracted from published data sets of humans receiving voriconazole intravenously (3 mg/ kg twice a day (bid)) or orally (200 mg bid) using XyExtract v2.5 (Wilton and Cleide Pereira da Silva, Campina Grande, Paraba, Brazil). Plasma concentration profiles for intravenous and or al dosing were best characterized by twocompartment models of zeroa nd first-order absorption, resp ectively (analysis by WinNonlin 5.2, Pharsight Corporation, Mountain View, CA). PK paramete rs for the two dosing regimens are presented in Table 2. These values were us ed to simulate the respective voriconazole plasma concentration-time profiles ( = 12 h; Scientist 3.0 software program) (Figure 4-4). The PK parameters for two dosing regimens were then applied to the Emax-model to simulate the expected in vivo kill curves, assuming plasma protein binding of 58% [8]. Simultaneous fitting of the identified model to the changing concentration kill-curves was performed in Scientist 3.0 non-linear least squares regression software program (MicroMath, Salt Lake City, UT) as previously described [102]. To determine the model that best fits the data for each isolate, graphs were visually inspecte d for quality of fit and several criteria were considered, such as model se lection criterion (MSC), coe fficient of determination (R2) and the correlation between measured and calculated data points. In the current study, we employed a dynamic in vitro model to determine time-kill curves of voriconazole against C. albicans reference strain, C. glabrata and C. parapsilosis bloodstream isolates (2 each) at .25, and 16MIC. The MICs were determined by Etest and the Clinical and Laboratory Standards Institute (formerly NCCLS) broth microdilution method [22, 46, 123], and were within twofold agreement (Table 2-1). We characterized vorico nazole-Candida interac tions using the dynamic in vitro system in which drug concentrations change over time in a manner consistent with the PK profile in

PAGE 69

69 humans [39, 102] (Table 3-2). PK parameters were extracted from published data sets of human receiving voriconazole using XyExtract v2.5 (W ilton and Cleide Pereira da Silva, Campina Grande, Paraba, Brazil) and the data were anal yzed by WinNonlin 5.2 (Pharsight Corporation, Mountain View, CA) to describe the best abso rption model. The Scientist 3.0 software program was then used to simulate plasma concentration-time profiles for multiple dosing of voriconazole ( = 12 h). Simulations of expected kill curves for each isolate were made using the PK parameters for two dosing regimens in the Emax-model as described above, assuming plasma protein binding of 58% [8]. As shown in Figure 4-4, the simulated kill curves predicted that voriconazole would exert prolonged fungistatic activity ag ainst all five isolates, resulting in diminishing Candidal concentrations over time. Indeed, for three isolates ( C. albicans ATCC90029, C. parapsilosis 1 and C. parapsilosis 2), reductions from starting inocula over 60 hours were predicted to exceed 2-logs. Of note, kill curves for four isolates were almost iden tical for the IV and oral dosing regimens. The exception was C. parapsilosis 2, for which the effects of IV dosing were predicated to be greater than oral dosing. Discussion Our study is noteworthy for severa l reasons. To our knowledge, this is the first report of time-kill experiments against Candida isolates using an in vitro method that mimics changing voriconazole concentrations in vivo rather than testing constant concentrations. The most striking finding from our study wa s that starting voriconazole con centrations of 1 MIC were largely ineffective at inhibiting the growth of is olates, whereas starting co ncentrations of both 4 and 16MIC significantly inhibited all isolates. The data from the dynamic model, therefore, suggest that clinicians sh ould aim for peak serum concentrations of voriconazole 4MIC in treating their patients with candidemia. Sin ce the median of maximu m voriconazole plasma

PAGE 70

70 concentrations in clinical trials was 3.79 g/ml [124], a target peak concentration 4MIC fits well with the recently proposed breakpoint MICs 1 g/mL for voriconazole susceptibility [122]. Second, we demonstrated that it is possible to accurately fit the data from our dynamic model using an adapted Emax mathematical model. Moreover, we used the mathematical model to simulate expected kill curves for typical do sing regimens of voriconazole by substituting the concentration term with PK data collected in vivo and accounting for protei n binding. In this regard, our approach of combining in vitro time-kill data with existing in vivo PK data might serve as a model for future stud ies to define the optimal use of antifungal agen ts. Although PD studies comparing the efficacy of various regimens of voriconazole and other antifungals are feasible in animal models [32], they are complicated, laborious and expensive. In vitro time-kills permit direct study of the inte raction between antifungals and fungi in a controlled and reproducible manner, and th ey allow direct comparisons of diffe rent agents and dosing strategies in a more convenient, faster and cheape r way that does not expend animal lives [126]. Finally, our data predict that typical IV a nd oral dosing regimens of voricon azole are comparable in their ability to kill most Candida isolates in vitro As such, the data lend support to the practice of switching to oral voricon azole to complete a course of th erapy of invasive candidiasis among patients who have exhibited clin ical and microbiologic responses to initial parenteral antifungal therapy. In conclusion, we have shown that the PK/P D relationship of voriconazole against Candida spp. can be accurately modeled in a dynamic in vitro system that mimics in vivo conditions. PK/PD modeling approaches such as ours merit further investigation as to ols to define optimal treatment regimens of voriconazole and other antifungal agents against candidiasis.

PAGE 71

71 Table 4-1. Voriconazole against Candida albicans ATCC90029 in changing concentration ex periments (Data of CFU/mL, Mean SD, n = 3) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 22.13E+055.98E+041.66E+054.04E+041.63E+053.32E+041.85E+053.86E+041.72E+055.24E+04 43.25E+057.81E+042.72E+055.40E+042.84E+052.16E+042.50E+053.81E+042.42E+058.13E+04 64.62E+058.61E+044.25E+058.52E+043.89E+054.60E+042.92E+054.01E+042.96E+056.86E+04 86.83E+051.51E+056.25E+057.77E+044.46E+056.18E+043.30E+057.16E+041.62E+052.58E+03 101.06E+061.67E+059.69E+051.08E+058.45E+059.47E+042.00E+051.71E+047.80E+043.92E+03 121.88E+062.36E+051.94E+065.03E+059.42E+057.51E+041.27E+053.16E+033.03E+044.35E+03 142.53E+065.20E+052.67E+066.74E+051.04E+061.85E+049.23E+046.31E+032.13E+046.21E+03 164.71E+061.62E+063.66E+061.54E+061.38E+062.71E+057.99E+041.39E+041.86E+044.58E+03 187.05E+063.85E+066.03E+062.85E+061.91E+063.47E+057.24E+041.83E+041.18E+042.54E+03 209.43E+061.68E+068.21E+061.83E+062.46E+062.00E+056.83E+041.41E+041.07E+041.99E+03 221.11E+072.89E+069.34E+062.91E+063.36E+063.72E+054.04E+049.00E+039.42E+031.10E+03 241.45E+071.48E+061.15E+077.47E+053.82E+064.91E+054.42E+045.10E+031.10E+042.96E+03

PAGE 72

72 Table 4-2. Voriconazole against Candida glabrata 1 in changing concentration experiment s (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.42E+051.67E+041.17E+052.07E+041.29E+052.37E+031.04E+051.04E+041.04E+057.22E+03 42.04E+051.58E+041.66E+051.06E+041.95E+052.68E+041.64E+055.69E+041.68E+054.36E+04 62.59E+055.81E+042.29E+058.03E+032.49E+058.48E+041.87E+055.75E+041.88E+054.00E+04 83.27E+054.91E+042.72E+055.58E+042.60E+058.17E+042.13E+056.09E+041.87E+051.67E+04 103.90E+051.17E+052.96E+054.97E+042.92E+051.07E+052.54E+057.58E+042.01E+053.16E+04 126.18E+052.36E+055.50E+051.47E+054.60E+051.73E+053.59E+057.56E+043.34E+057.96E+04 149.20E+059.91E+041.01E+064.91E+047.20E+054.30E+043.58E+051.45E+043.65E+058.58E+04 161.32E+062.05E+041.13E+068.66E+048.32E+053.31E+045.05E+051.15E+054.40E+059.53E+02 181.90E+062.96E+041.48E+061.88E+051.06E+062.52E+045.69E+056.91E+044.46E+051.37E+03 202.66E+062.28E+051.60E+062.40E+041.19E+069.47E+045.57E+055.27E+044.64E+054.67E+02 224.03E+068.75E+052.37E+063.97E+051.70E+062.11E+045.38E+053.66E+044.01E+051.39E+04 244.13E+067.98E+052.97E+066.44E+051.76E+065.27E+045.60E+051.68E+043.89E+051.40E+04

PAGE 73

73 Table 4-3. Voriconazole against Candida glabrata 2 in changing concentration experiment s (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.57E+055.31E+041.31E+051.09E+041.41E+052.86E+031.10E+051.18E+041.21E+059.25E+02 42.21E+051.58E+041.79E+051.08E+042.27E+052.20E+041.82E+057.21E+031.69E+054.62E+04 63.07E+057.51E+042.61E+053.51E+043.79E+058.15E+042.42E+058.49E+042.07E+055.75E+04 84.35E+051.20E+053.84E+052.51E+044.21E+059.94E+043.20E+059.63E+042.40E+051.87E+04 105.43E+058.13E+045.41E+051.96E+035.96E+051.72E+054.04E+059.96E+043.06E+054.68E+04 129.80E+054.67E+057.28E+057.96E+046.89E+051.70E+054.28E+057.68E+043.33E+054.29E+04 141.24E+062.20E+051.07E+065.29E+018.48E+052.28E+054.48E+058.55E+043.53E+053.85E+04 162.22E+068.49E+051.81E+063.12E+051.50E+064.50E+057.17E+052.06E+054.55E+051.12E+04 183.54E+061.21E+062.80E+061.03E+062.52E+069.34E+057.77E+051.56E+055.06E+051.37E+05 204.40E+061.27E+063.42E+068.45E+053.11E+068.76E+058.36E+051.72E+054.93E+059.74E+04 225.80E+062.40E+064.87E+061.16E+063.30E+068.54E+057.65E+051.87E+054.45E+057.21E+04 245.95E+061.77E+065.54E+061.46E+063.42E+068.15E+057.89E+052.01E+054.32E+053.21E+04

PAGE 74

74 Table 4-4. Voriconazole against Candida pa rapsilosis 1 in changing conc entration experiments (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.86E+057.74E+041.89E+051.03E+042.06E+054.45E+041.80E+054.99E+041.72E+054.79E+04 44.80E+052.20E+053.66E+052.91E+043.87E+053.68E+044.18E+052.75E+042.73E+059.65E+04 68.96E+053.56E+056.49E+054.19E+046.16E+054.00E+034.96E+059.34E+043.87E+051.77E+05 81.70E+064.70E+051.46E+061.08E+059.61E+053.02E+051.88E+053.16E+041.64E+057.10E+04 102.43E+064.01E+052.30E+064.47E+051.60E+063.22E+051.44E+056.45E+037.09E+041.95E+04 123.60E+061.38E+063.63E+062.02E+063.02E+061.29E+061.11E+052.48E+043.27E+048.60E+03 145.88E+061.88E+065.35E+062.58E+054.71E+062.37E+061.37E+051.23E+044.42E+041.74E+04 168.48E+062.81E+068.55E+061.99E+056.44E+062.76E+061.71E+052.10E+045.59E+042.58E+04 181.06E+072.20E+061.21E+072.89E+068.27E+062.81E+061.64E+054.07E+044.54E+042.31E+04 201.54E+072.45E+061.43E+071.46E+061.13E+072.20E+062.16E+059.19E+035.28E+042.01E+04 221.89E+074.79E+061.27E+071.17E+051.11E+073.23E+063.20E+058.26E+045.73E+042.71E+04 242.27E+073.06E+051.67E+077.17E+061.37E+073.39E+065.32E+052.16E+055.95E+042.56E+04

PAGE 75

75 Table 4-5. Voriconazole against Candida pa rapsilosis 2 in changing conc entration experiments (Data of CFU/mL, Mean SD, n = 2) Time (h)Control SD0.25MICSD1MICSD4MICSD16MICSD 01.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+001.00E+050.00E+00 21.85E+053.82E+041.63E+054.84E+041.35E+054.03E+031.84E+058.69E+041.33E+057.26E+04 42.58E+053.30E+042.75E+054.05E+042.37E+055.48E+041.19E+053.61E+045.66E+044.88E+03 63.80E+051.86E+053.07E+051.21E+053.60E+051.76E+051.06E+054.04E+045.40E+041.47E+04 85.20E+052.29E+056.18E+051.21E+056.14E+052.69E+051.07E+053.22E+042.86E+043.83E+03 101.62E+064.52E+051.59E+062.18E+041.23E+063.62E+058.58E+045.19E+032.22E+041.35E+04 122.32E+063.79E+042.51E+064.20E+051.87E+062.45E+058.59E+044.97E+041.22E+046.37E+03 142.23E+063.06E+052.74E+063.13E+042.97E+064.30E+039.11E+044.51E+041.17E+041.35E+03 163.09E+065.82E+052.96E+065.50E+052.86E+062.18E+051.67E+059.81E+031.44E+042.70E+02 184.16E+061.28E+065.42E+062.64E+064.25E+061.47E+062.97E+055.19E+041.16E+044.84E+03 204.57E+061.12E+065.23E+061.25E+054.19E+061.05E+063.50E+055.63E+049.13E+033.80E+03 225.23E+064.25E+055.47E+068.45E+054.86E+066.69E+054.97E+052.07E+051.63E+047.84E+03 245.73E+061.41E+066.83E+061.86E+063.79E+064.23E+057.40E+053.68E+051.29E+046.77E+03

PAGE 76

76 Table 4-6. Pharmacodynamic parameters and goodness of fit criteria against Candida isolates in the dynamic infection model Parameter (unit) C. abicans ATCC90029 C. glabrata 1 C. glabrata 2 C. parapsilosis 1 C. parapsilosis 2 C (mg/L)a0.002/0.008/0.032/0.1280.0475/0.19/0.76/3.040.008/0.032/0.128/0.5120.002/0.008/0.032/0.1280.004/0.016/0.064/0.256 Ks (h-1) 0.29 0.17 0.19 0.32 0.30 Kmax (h-1) 0.45 0.27 0.38 0.49 0.50 EC50 (mg/L) 0.0050.0740.08100.0070.017 1.21 5.55 6.98 9.81 9.34 0.556 0.061 0.085 0.25 0.70 Nmax (CFU/mL) 1.65E+07 1.87E+07 2.48E+07 1.96E+07 5.88E+06 h 1.42 0.86 0.89 1.79 2.86 MSC/R23.58/0.98 3.68/0.99 2.71/0.97 3.19/0.98 3.08/0.98 Cmax: maximum voriconazole concentration ks: fungal growth rate constant in the absence of voriconazole kmax: maximum killing rate constant (maximum effect) EC50: concentration of voriconazole necessary to produce 50% of maximum effect : constant used to fit the ini tial lag phase for the growth : constant used to fit the initial la g phase for the inhi bition or killing h: Hill factor MSC: model selection criteria a Values of voriconazole concentrations of 0.25, 1, 4and 16MIC, respectively.

PAGE 77

77 )1(0 tkeeVV Figure 4-1. Concentration elimination curve of voricona zole in the dynamic infection model C. parapsilosis 1y = 0.0087e-0.1144xR2 = 0.9931 0.000 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 024681012141618202224 Time(h)Vori conc. (ug/ml) C. parapsilosis 2y = 0.0174e-0.1141xR2 = 0.9929 0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 024681012141618202224 Time (h)Vori conc. (ug/ml)

PAGE 78

78 C. glabrata 1 0 4 8 12 16 20 24 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.0475mg/L N3:0.19mg/L N4:0.76mg/L N5:3.04mg/L C. glabrata 2 0 4 8 12 16 20 24 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.008mg/L N3:0.032mg/L N4:0.128mg/L N5:0.512mg/L C. parapsilosis 1 0481 21 62 02 4 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L C. parapsilosis 2 0481 21 62 02 4 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 N1:control N2:0.004mg/L N3:0.016mg/L N4:0.064mg/L N5:0.256mg/L C. albicans ATCC90029Time_hr 0481 21 62 02 4 N_CFU/mL 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L Figure 4-2. Time-kill curves from the dynamic in vitro model (Mean SD; n=3, without significant differences in results by ANOVA). Voriconazole concentrations were 0, 0.25, 1, 4 and 16MICs (listed as N1 -N5 in the individual figure legends).

PAGE 79

79 C. albicans ATCC90029Time_hr 0481 21 62 02 4 N_CFU/mL 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 1 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.0475mg/L N3:0.19mg/L N4:0.76mg/L N5:3.04mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 2 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.008mg/L N3:0.032mg/L N4:0.128mg/L N5:0.512mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 2 0481 21 62 02 4 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.004mg/L N3:0.016mg/L N4:0.064mg/L N5:0.256mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 1 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 1 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.0475mg/L N3:0.19mg/L N4:0.76mg/L N5:3.04mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 2 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.008mg/L N3:0.032mg/L N4:0.128mg/L N5:0.512mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 2 0481 21 62 02 4 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.004mg/L N3:0.016mg/L N4:0.064mg/L N5:0.256mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 1 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.0475mg/L N3:0.19mg/L N4:0.76mg/L N5:3.04mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. glabrata 2 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.008mg/L N3:0.032mg/L N4:0.128mg/L N5:0.512mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 2 0481 21 62 02 4 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 N1:control N2:0.004mg/L N3:0.016mg/L N4:0.064mg/L N5:0.256mg/L N1_cal N2_cal N3_cal N4_cal N5_cal C. parapsilosis 1 04812162024 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 N1:control N2:0.002mg/L N3:0.008mg/L N4:0.032mg/L N5:0.128mg/L N1_cal N2_cal N3_cal N4_cal N5_cal Figure 4-3. Fitted time-kill curves were deri ved by our mathematical model for changing concentrations of voriconazole (Mean SD; n=3, without significant differences in results by ANOVA). Voriconazole concen trations were 0, 0.25, 1, 4 and 16MICs (listed as N1-N5 in the individual figure legends).

PAGE 80

80 C. glabrata 1 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. glabrata 2 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. albicans ATCC90029Time_hr 01 02 03 04 05 06 0 N_CFU/mL 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 ctrl growth_CFU/mL oral kill_CFU/mL IV kill_CFU/mL Ct_oral_mg/L Cf_oral_mg/L Ct_iv_mg/L Cf_iv_mg/L C. parapsilosis 1 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. parapsilosis 2 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. glabrata 1 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. glabrata 2 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. albicans ATCC90029Time_hr 01 02 03 04 05 06 0 N_CFU/mL 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 ctrl growth_CFU/mL oral kill_CFU/mL IV kill_CFU/mL Ct_oral_mg/L Cf_oral_mg/L Ct_iv_mg/L Cf_iv_mg/L C. parapsilosis 1 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 C. parapsilosis 2 01 02 03 04 05 06 0 1e-1 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 Figure 4-4. Using parameters from dynamic models to simulate candidal time-kills and plasma voriconazole concentration-time profiles: The simulations of time-kills are shown for short-term IV infusion () a nd oral dosing of voriconazole (---). Control growth is in the absence of drug (). The simulated plasma concentration-time profiles for multiple doses of voriconazole ( =12h) are shown for short-term intravenous (IV) infusion (T=1h, infusion rate=3mg/kg/h, bid); and oral administration (200mg, bid, F=0.9). Free voriconazole concentrations were calculated assu ming plasma protein binding level of 58%.

PAGE 81

81 CHAPTER 5 IN VITRO MICRODIALYSIS OF VORICONAZOLE Background Microdialysis is a dynamic technique which has been employed for the in vivo measurement of a variety of drugs and endogenous compounds in different tissues [73, 99]. The applicability of this technique can be limited by drug lipophilic ity which can impair the diffusion through microdialysis semi-permeable membrane [130, 131]. Voriconazole is a moderately lipophilic (Log D7.4 = 1.8) antifungal triazolic agent [31]. The determination of the recovery (the fraction of the free drug at the site of sampli ng which dialyses into the probe) is essential for voriconazole to estim ate the free drug levels in the tissue. There are several factors which affect drugs rela tive recovery including the drugs physicochemical properties, flow rate, probes charac teristics such as type membrane length and diameter, experimental conditions such as temperature and matrix tortuosity of the interested tissues, when the recovery is determined in vivo [99]. For determining the in vitro recovery of voriconazole in microdialysis, a straight -forward approach was undertaken and the calibration me thods directly were used to determine the free, unbound drug concentration. For practical reasons, the two mo st common methods are the extraction efficiency method (EE) and retrodialysis (RD) method. Specific Aims The objective of in vitro microdialysis (MD) study was to assess the feasibility of doing microdialysis of voriconazole, to determine the in vitro recovery of this compound and to determine the value of protein bi nding of voriconazole in both rat and human plasma in vitro

PAGE 82

82 Materials and Methods Materials All materials are listed in Tabl e 5-1 unless otherwise stated. The CMA/20 probe, with a membrane lengt h of 10 mm and a molecular cutoff of 20 kDa was used for the experiments. The probe was connected to a 1000 l Gastight syringe by a Perifix screw connector (B. Braun). A microinfus ion pump (Harvard Apparatus Syringe Pump Model ) was us ed to keep the flow through the probe constant. All the solvents used in the HPLC analysis were of HPLC grade. Methods Recovery The in vitro recovery of voriconazole was de termined by two different methods: extraction efficiency (EE) and retrodialysis (RD). All methods were carried out at 37C. Each procedure is describe d in the following sections. Extraction efficiency method (EE) In the EE method, blank Lactated Ringers (LR) solution was pumped through the microdialysis (MD) probe at a flow rate of 1.5L/min. The MD probe was then placed into the sample tube, which was a glass t ube filled with approximately 6 ml of drug solution, starting with the lowest concentrati on and fixed in place with adhesive tape. Six different voriconazole concentratio ns of 0.5, 1, 2, 5, 10 and 40 g/ml were used in this experiment with all prepared in lactated Ringers solution. To guarantee equal concentrations throughout the whole tube, the solution was stirred at approximately 100rpm. Voriconazole was diffused from the sa mple tube into the MD probe and were measured in the dialysate. Samples were colle cted every 20 minutes after the end of the 2

PAGE 83

83 hours equilibration time. A total of 5 sample s were collected for each experiment. The concentrations were determined by a valid ated HPLC/UV method and plotted versus time. All experiments were performed in trip licates. The percent recovery (R%) for the EE method is then calculated as followed: 100 C %dialysatesolC R (5-1) Where R% is the recovery in percentage, Csol is the average voricona zole concentration in the tube before and after the experiment, and Cdialysate is the voriconazole concentration in the dialysate. Retrodialysis method (RD) In RD method, the syringes contained th e voriconazole solution that was pumped through the MD probe at a flow rate of 1.5L/min. The MD probe was placed into a sample tube that was filled with blank LR so lution and fixed in place with adhesive tape. To guarantee equal concentra tions throughout the w hole tube, the solution was stirred at approximately 100rpm. The voriconazole was di ffused out of the probe into the sample tube. The loss of voriconazole over the membrane was then determined while determining Cdialysate. Samples are collected every 20 mi nutes after the end of the 2 hours equilibration time. A total of 5 samples were collected for each experiment. The voriconazole concentrations were determ ined by a validated HPLC/UV method and plotted versus time. All experiments were performed in triplicates. The percent recovery (R%) for the RD method was then calculated as followed: 100 )C-(C %dialysate perfusate perfusateC R (5-2)

PAGE 84

84 Cperfusate is the average voriconazole concentra tion in the perfusate before and after the experiment, and Cdialysate is the voriconazole concen tration in the dialysate. Microdialysis experiments The in vitro recovery of the voriconazole was determined by the extraction efficiency method (EE). In this method, bl ank Ringers solution was pumped through the microdialysis probe (CMA/20), which was pl aced into a testing tube filled with approximately 6 ml of voriconazole solution. Six different voriconazole concentrations of 0.5, 1, 2, 5, 10 and 40 g/ml were tested in this experime nt with all prepared in lactated Ringers solution. After placing the probe in the drug solution, it was allowed to equilibrate for 10 minutes at flow rate of 5 l/min followed by 2 hours at 1.5 l/min. Subsequently dialysate samples were collected every 20 minutes after the equilibration period. A total of 5 samples were collected for each experiment, and the experiment was triplicate for each concentration. The voriconazole concentration in the dialysates and in the tube before and after the experime nts was determined by the HPLC/UV method described above. The probe recovery dete rmined by the extraction efficiency was calculated by the equation 5-1. The protei n binding was determined by the extraction efficiency was calculated by the equation 5-3. 100) % C 1(%dialysate solCR binding protein (5-3) HPLC for determination of voriconazole An HPLC protocol for measuring voricon azole in Lactated Ringers solution and human or rat plasma was developed based an existing assay method with a calibration range of 0.025 g/ml voriconazole as being described previously in Materials and Methods of Chapter 2. Voriconazole samp les in Lactated Ringers solution were

PAGE 85

85 measured using HPLC/UV method directly. Pl asma samples were extracted before the measurement using HPLC /UV method as being de scribed previously [120] with modification. Briefly, plasma sa mples were centrifuged at 1,700 g for 5 min. An aliquot (150 l) was pipetted into a 1.5 ml Eppendor f tube and acetonitrile (240 l) was then added. The mixture was mixed brie fly by vortex, centrifuged at 1,200 g for 5 min after standing for 10 min at room temperature a nd supernatant was tr ansferred to a new Eppendorf tube. This step was repeated one more time. Then 50 l of the supernatant plus 50 l of 0.04 M Ammonium Phosphate buffer (pH 6.0) was applied for HPLC system. Calibration curves for voriconazole Stock solutions of voriconazo le (1 g/ l) were prepar ed in distilled water and diluted in Lactated Ringers solution or plasma to give 100 g/ ml. Standards were prepared by adding the diluted voriconazole so lution to appropriate volumes of Lactated Ringers solution or plasma to give a concentrati ons of 0.025, 0.1, 0.2, 0.5, 1, 2, 4, 8 and 16 g/ ml. The plasma standards were also extracted before injecting to HPLC system. Calibration curves were constructed by plotting the peak ar ea of voriconazole against concentration using a weighted (1/X2) least squares regression for HPLC data analysis. Results The HPLC data printout from the inte grator was transferred into EXCEL 2003 (Microsoft Corporation) spreadsheets. With the equation obtained from the Weighted (1:X2) Linear Regression, the voriconazole concentrations rom Lactated Ringers solution, rat and human plasma samples were calculated [132].

PAGE 86

86 Stability of Voriconazole in Lact ate Ringers Solution and Plasma The in vitro stability of voriconazole in Lact ate Ringers solution and human and rat plasma were studied. The results show ed that voriconazole in Lactate Ringers solution, human and rat plasma were found stable at -80C for at least 6 months, at -4C at least for 72 hours, room temperature and 37 C for at least 24 hours which is consistent with previous report [4, 30]. Protein Binding The results of recovery for both extraction efficiency (EE) and retrodialysis (RD) methods were summarized in Table 5-2 a nd 5-3. As shown from the tables, the in vitro recovery of this compound from both extr action efficiency (EE) and retrodialysis methods (RD) were quite similar with 51.1% 2.6% (mean SD) (ranging from 46.2% to 56.3%) for EE method and 51.9% 2.9% (mean SD) (ranging from 45.3% to 57.8%) for RD method. The results of protei n binding of voriconazole in presence of different drug concentrations in rat and huma n plasma were summari zed in Table 5-4 and 5-5. The protein binding of voriconazole in rat plasma is 66.2% 2.2% (mean SD) (ranging from 61.5% to 70.3%) and that in human plasma is 57.0% 2.8% (mean SD) (ranging from 50.9% to 62.1%). Discussion This experiment confirmed that the perf ormance of voriconazole microdialysis was feasible. It has been reported in the literature that rec overies were dependent on the method used for the determination as well as on the flow rate, but independent of drug concentration [133]. In our current study, the recoveri es of voriconazole were independent of both the methods (EE and RD ) and drug concentrations.

PAGE 87

87 Only the free or unbound fraction of drug is available for antimicrobial activity according to the free drug hypothesis of anti-infective agents. Voriconazole exhibits moderate plasma protein binding with values of 66% in rat plasma and 58% in human plasma, which was measured by 14C Voriconazole [31]. The binding is independent of dose or plasma drug concentrations [8]. Our current study is the first time to report the voriconazole protein binding with in vitro microdialysis method as our knowledge. The average plasma protein binding of voriconazole in rat and human is 66.2% and 57.0%, respectively, which is consistent with the previous publications [8, 31]. Moreover, we also confirmed that the protein binding in rat a nd human plasma was independent of the drug concentrations. The free drug concentrations provided a better correlation to outcome than total drug concentrati ons. Free drug concentrations must be considered when examining the relationship between pharmacokinetic parameters and in vivo activity. Microdialysis provided a simple, clea n method for determining the free drug concentration in vitro and in vivo

PAGE 88

88 Table 5-1. List of materials Materials Resource Voriconazole Pfizer, New York, NY Acetonitrile Fisher, HPLC grade, Pittsburgh, PA Ammonium Phosphate Sigma, HPLC grade, St. Louis, MO Lactated Ringer's solution (USP)Abbott, Chicago, IL Rat plasma Wistar, Harlan Sprague-Dawley, Indianapolis, IN Human plasma Civitan Regional blood system, Gainesv ille, FL CMA/20 probe Stockholm, Sweden Balance AB104 Mettler, Toledo, Hightstown, NJ Vortex Kraft Apparatus model PV-5, Fisher, Pittsburgh, PA Pipette tips Fisher, Pittsburgh, PA Tubes microcentrifuge Fisher, Pittsburgh, PA Centrifuge Fisher model 235V, Pittsburgh, PA

PAGE 89

89 Table 5-2. Extraction efficiency (EE) method for measur ing voriconazole recovery Recovery Concentration (g/ml) Average (%)0.511.042.065.1310.3141.47 150.650.746.856.352.650.6 246.651.853.452.854.446.1 346.252.052.251.556.249.6 Mean (%)47.851.550.853.554.448.851.1 SD (%)2.40.73.52.41.82.32.6 Table 5-3. Retrodialysis (RD) method for measuring voriconazole recovery Recovery Concentration (g/ml) Average (%)0.520.991.965.1210.3940.37 149.951.649.456.257.847.6 247.151.853.552.355.751.5 345.353.253.754.352.751.2 Mean (%)47.452.252.254.355.450.151.9 SD (%)2.30.92.51.92.62.22.9

PAGE 90

90 Table 5-4. Rat plasma pr otein binding data of voriconazole measured by in vitro microdialysis P rotein bindin g Concentration (g/ml) Average (%)0.531.132.225.4211.0941.19 161.564.966.065.866.970.3 262.466.768.568.569.263.2 361.769.163.368.468.765.7 Mean (%)61.966.965.967.568.266.466.2 SD (%)0.42.12.61.51.23.62.2 Table 5-5. Human plasma protein binding data of voriconazole measured by in vitro microdialysis P rotein bindin g Concentration (g/ml) Average (%)0.541.092.165.5511.1541.13 150.951.856.557.661.262.1 255.955.956.660.260.860.3 356.854.451.559.855.957.9 Mean (%)54.654.054.959.259.360.157.0 SD (%)3.22.12.91.42.92.12.8

PAGE 91

91 CHAPTER 6 IN VIVO PHARMACOKINETIC STUDIES OF VORICONAZOLE IN RATS Background Voriconazole is an antifunga l agent which has been show n to be effective in the treatment of patients with candidiasis [134-136]. The effectiveness of voriconazole and other antifungal agents is determined by the sensi tivity of the fungi a nd the concentrations achieved at the site of infection [34, 131, 137]. However, the site of infection is generally not the blood. Therefore, in many infections con centrations in organ tissues, extracellular fluids, and inflammatory cells rather than pl asma concentrations determine the clinical outcome and may allow a better prediction of the therapeutic effect than plasma concentrations [73, 74, 95, 99, 138-141]. Microdialysis has been used to study drug free levels in many tissues and organs, such as muscle, adi pose tissue, bile, blood, eye, skin, brain, lung, and so on [73, 74]. It is an established tech nique for studying physiological, pharmacological, and pathological changes of many drugs such as antibiotics [142, 143], anti-cancer drugs [95, 144-146], and psychoactive compounds [137, 147-149] in both preclinical and clinical studies. Specific Aims The objective of this project is to study the distribution profile of voriconazole in the body by measuring its concentration in pl asma, muscle. To evaluate unbound muscle concentrations of voriconazole in Wistar rats by microdialysis after intravenous administration of voriconazole and to comp are the free tissue concentration to free plasma concentration. To develop a populati on PK model of voricona zole in rats using NONMEM.

PAGE 92

92 Materials and Methods Reagents and Equipment All of them have been described in Table 5-1 in Chapter 5 unless otherwise stated. Isoflurane, USP was distributed by Webste r Veterinary. Supply, Charlotte, NC Animals Wistar male rats (Harlan Sprague-Dawley, Indianapolis, IN) were used in the project. Previous studies with other azole drugs have indicated that rat is a suitable test model to investigate tissue distribution using microdialysis [133, 137, 147, 148]. The rats were shipped to ACS a couple of days before the study starts in order to reduce stress and adapt them to the researcher. In the experime nt, the rats were weighed before the surgery and dose administration. The body weight of male Wistar rats used in this experiment was ranged from 350g to 400g and age of 2-4 m onths. The animals were numbered in the sequence of the experiments wit hout identifying devices such as tattoos or collar numbers since this project i nvolved non-survival su rgical experiment. Experimental Design Two groups of anesthetized male Wistar rats of six each were administered intravenously with either 5mg/kg or 10 mg/kg dose of voriconazole. Total plasma concentrations as well as unbound concentrations in thigh muscle were measured over 8hour period by microdialysis. Animals were sacrif iced at the end of the study. The data were analyzed by both non-compartmental and compartmental pharmacokinetic approaches. The study was performed with the approval of the Institutional Animal Care and Use Committee (IACUC) of the University of Florida (Protocol # F109). The animals

PAGE 93

93 were housed in the Animal Care Services (ACS) at the Health Science Center, University of Florida according to the st andard husbandry procedures. Anesthetic Procedure Isoflurane has been used in the surgical procedure for anesthesia in rats in many reports [150-153]. In our studies, the rats were anesth etized by using Isotec-4 isoflurane vaporizer (SurgiVet/ Smith s Medical, Waukesha, WI) from ACS. After placing the rat in the induction chamber, the flowmeter was adju sted to 0.8-1.5L/min and the isoflurane vaporizer was adjusted to 2% which was mark ed as Anesthesia was confirmed by the absence of reflexes after pinching the rats footpads. After anesthesia, the rat was immobilized in a supine positi on on a dissecting board by hol ding the paws with rubber bands, which was attached to pins on each corner of the board. For maintenance of anesthesia, a mask connected to the Bain Circuit was used for the rat, and the flowmeter was adjusted to 400-800mL/min. The footpads were pinched in a regular basis (every 15 min) during the experiment. The rat was kept normothermic on an electric heating pad. Antiseptic Procedure This project involved non-su rvival surgical experiment The surgery sites were disinfected by swiping the area with 70% isopropyl alcohol and skin areas of surgery sites were shaved before surgery. Blood Samples Blood sample were collected by a polyethyle ne catheter (inner diameter of 0.3 mm and outer diameter of 0.7 mm) introduced in the carotid. After initial insertion and following each sampling, the catheter was irrigated with 300 l 100 IU pre-warmed heparinized saline. Blood samples (100 l) were collected in heparinized tubes before the dose was administered (time zero) and at 10, 30, 60, 90, 120, 180, 240, 300, 360, 480min

PAGE 94

94 after drug administration. The blood sample s were centrifuged at approximately 2000 rpm for 10 minutes at room temperature and th e supernatant of plasma was transferred to a new Eppendorf tube. This step was repeated one more time. Then plasma samples were kept at -80 C until analyzed. Microdialysis Probe calibration The left hind leg muscle was used for inse rtion of a microdialys is probe after skin area was shaved and cut open. Before introducin g the probe itself, a guide plastic cannula was introduced in the muscle with the help of a 23-guage needle. Afterwards, the plastic cannula was kept in place and the needle wa s removed and replaced by the microdialysis probe. After inserting the probe the plastic cannula was re moved by tearing it out while holding the probe in place. Probe calibration was performed before drug administration. The probe was initially perfused with lactated Ringe rs solution (NaCl 137 mM, KCl 1.0 mM, CaCl2 0.9 mM, NaHCO3 1.2 mM) at a flow rate of 2 l /min for 30 minutes using a Harvard Apparatus 22 injection pump, model 55-4150. After equilibration, the probe was calibrated by retrodialysis. In this method, the syringe with lactated Ringe rs solution was replaced by a syringe with a vor iconazole solution of 1 or 2 g/ml. The drug solution was initially pumped at a flow of 5 l/min for 5 min and after the flow was changed back to 2 l/min. The probe was allowed to equilibrate for 0.5 hour before samples were collected. The samples for the calibration procedure were collected very 20 minutes and a total of 3 samples were collected in each experiment. The samples were frozen at -80 oC before analyses by a validat ed HPLC/UV method.

PAGE 95

95 Muscle microdialysis After the probe calibration, the drug perfusion was stoppe d and removed instead of blank lactated Ringer's solution. The probe was perfused with lactated Ringer's solution for 30 minutes to wash out. After wash out microdialysis samples were collected for muscle over 20-minute intervals at times 20, 40, 60, 80, and up to 480 minutes after drug administration. The in vivo recovery was calculated by the same above equation of 100 C C C %perfusate dialysate perfusate R (6-1) The drug concentration in the tissue wa s calculated using the equation of % Cdialysate ,R Cfreet (6-2) Ct,free is the muscle free concentration, Cdialysate is the concentration in the dialysate, and R% is the percent of recovery obtained in each animal experiment. Data Analysis Microdialysis samples were measured using HPLC/UV method directly. Plasma samples were extracted before the meas urement using HPLC/ UV method as being described previously (Methods in Chapter 5) with modification. Brie fly, plasma samples were thawed on ice and then centrifuged at 1,700 g for 5 min. An aliquot (25 l) was pipetted into a 0.7 ml Eppendorf tube and acetonitrile (40 l) was then added. The mixture was mixed briefly by vortex, centrifuged at 1,200 g for 5 min after standing for 10 min at room temperature and supernatan t was transferred to a new Eppendorf tube. This step was repeated one more time. Then 25 l of the supernatan t plus 25 l of 0.04 M Ammonium Phosphate buffer (pH 6.0) was applied for HPLC system.

PAGE 96

96 The HPLC data was input into EXCEL 2003 (Microsoft Corporation) spreadsheets. With the equation obtained from the Weighted (1:X2) Linear Regression, the rat and human plasma sample concentrations were calculated [132]. With the equation obtained from the Weighted (1:X2) Linear Regression function, the rat plasma sample concentrations were calculated. Unbound concentrations in rat thigh muscle were measured by microdialysis. I ndividual probe recovery was determined by retrodialysis and allowed conversion of the measured dial ysate concentrations to the actual unbound tissue concentrations. The data were analyzed both by noncompartmental and compartmental pharmacokinetic approaches. Noncompartmental pharm acokinetic analysis The following parameters were determined for each rat for both plasma and microdialysis noncompartmental pharmacokinetic analysis. The mean and standard deviation (SD) of each parameter were also determined. The data were analyzed by WinNonlin 5.2 (Pharsight Corporation, Mountain View, CA) to describe the best absorption model. The initial concentration C0 was determined by logarithmic back-extrapolation to t = 0 using the first two data points. The terminal elimination rate constant (ke) was determined by linear regression of the log plasma concentrations. The terminal half-life (t1/2) was calculated as ln(2)/ke. The area under the concentration-time curve (AUCt) values were calculated between time 0 and the final time point at which measurable dr ug concentrations were observed, using the linear trapezoidal rule. AUC was calculated by extrapolation from time zero to infinity with ke. The area under the first moment curve (A UMC) was calculated from a plot of Ct versus t using the trapezoidal rule up to the last data point (Cx) at time tx and adding the extrapolated terminal area, calculated as Cxtx/ke + Cx/ke2. AUMC was Area under the

PAGE 97

97 first moment curve when the time concentra tion curve is extrapolated to infinity. The mean residence time (MRT) was calculated as AUMC/AUC. Clearance (CL) was calculated by the relationship of Dose/AUC and the volume of distribution was calculated by the relationship CL/ k e. The volume of distribu tion at steady state (Vdss) was calculated as (DoseAUMC)/AUC2. Compartmental pharmacokinetic analysis and modeling Both the individual and average plasma and free tissue conc entrations at each time point were fitted by using the modeling software of NONMEM (UCSF, San Francisco, CA). Rat total plasma voriconazole pharmacokinetic data were first analyzed using oneand two-compartment model with linear or nonlinear elimination (Michaelis-Menten elimination). Then rat total plasma and unbound muscle voriconazole data were analyzed simultaneously using a two-compartment m odel with nonlinear elimination. The schemes of oneand two-compartment model with nonlinear elimination were shown in Figure 6-1 and Figure 6-2. The models were fitt ed to the data using the first order (FO) method and the subroutine ADVAN1 TRANS2 (one-compartment with linear elimination), ADVAN3 TRANS4 (two-compartment with linear elimination), ADVAN10 TRANS1 (one-compartment with nonlinear elimination), and ADVAN6 TRANS1 (two-compartment with nonlinear elimination). NONMEM performs linearized maximum likelihood estimation by use of an objective function (OF). Typical values (population means) with their corresponding standard errors (SE) and the intersubject and intrasubject variabilities were expressed as co efficients of variati on (CV percentage). To determine whether there was a statistically significant diffe rence between the goodness of fit between the two models, the Akaike model selection criteria wa s used, which required

PAGE 98

98 a decrease of two points in the objective function (minus twice the logarithm of the likelihood of the model) to accep t a model with one additional parameter, as well as a comparison of diagnostic plots (observed con centrations compared with predictions and observations/predictions compared with time). The one-compartment model with linear elimination will be parameterized in terms of CL (clearance), and V (volume of central compartment). The two-compartment model with linear elimination will be parameterized in terms of CL (clearance), V1 (volume of central compartment), V2 (volume of periphera l compartment), and Q (intercompartmental clearance). The one-compartment model with nonlinear elimination will be parameterized in terms of Vmax (maximum elimination rate), Km (the Michaelis-Menten constant, voriconazole concentration at which the elimin ation is at half ma ximum) and V (volume of central compartment). The two-compartmen t model with nonlinear elimination will be parameterized in terms of Vmax (maximum elimination rate), Km (the Michaelis-Menten constant, voriconazole concentration at which the elimination is at half maximum), V1 (volume of central compartment), V2 (volume of periphera l compartment), and Q (intercompartmental clearance). The differential equations for a two-compartment model with nonlinear elimination (Michaelis-Menten elimination) are given in following equations: 1 )1( 1 V A C (6-3) )1( )2( )1(12 1 1max 21Ak CK CV Ak dt dAm (6-4)

PAGE 99

99 )2()1( )2(21 12AkAk d t dA (6-5) 1 12 V Q K (6-6) 2 21V Q K (6-7) Where A(1) represents the amount of vor iconazole in the central compartment, A(2) represents the amount of voriconazole in the peripheral compartment, Vmax represents the maximal elimination rate, Km is the Michaelis-Menten constant (voriconazole concentration at which the elimination is at half maximum), C1 is the predicted serum concentrat ion of voriconazole, and k12 and k21 are the intercompartmental rate constants. V1 is the volume of central compartment, V2 is the volume of peripheral compartment, and Q is the intercompartmental clearance. Because free rather than protein-bound vor iconazole was measured in dialysate, a scaling factor for the tissue compartmen ts was introduced, accounting for the free fraction of voriconazole, f u fu V A C 2 )2( 2 (6-8) Interindividual variability for the pha rmacokinetic parameters was modeled using the following exponential error model: )(1 1max EXP V (6-9) )(2 2 EXP Km (6-10) )(3 31 EXP V (6-11) )(4 4 EXP Q (6-12)

PAGE 100

100 )(5 52 EXP V (6-13) Where is the population mean estimate (o r typical value) of the corresponding pharmacokinetic parameter, and is are the associated interindividual variability. The values are independent, identic ally distributed random erro rs with mean of zero and a variance equal to i2. Interindividual variability will be described by an exponential error model. When analysising the total plasma only, we used a proportional error model for the intraindividual residual variability of plasma concentration for estimation, which was described by the following equation: )1( ~ propij ij ijpCCp (6-14) where Cpij is the observed value of the jth plasma concentration of individual i; pij is the predicted jth plasma concentration of individual i; and ij, prop denote the proportional residual random errors distributed with zero means and variance 2 prop. When analysising the total plasma and unbound muscle voriconazole data simultaneously, the intraindividual residual variability of total plasma concentration were estimated using a proportional model, as de scribed previously and the unbound muscle data were using a combination of proportional and additive error model, as described by the following equation: addij propij ij ijpCCp, ,)1( ~ (6-15) where Cpij is the observed value of the jth plasma concentration of individual i; pij is the predicted jth plasma concentration of individual i; and ij, prop and ij, add denote the proportional and additive resi dual random errors distri buted with zero means and variance 2 prop and 2 add.

PAGE 101

101 Plasma total voriconazole concentrat ions and unbound muscle voriconazole concentrations were assigned separate residual errors, 1, 2 and 3, respectively. The following criteria were considered to determine the best model: In the case of nested models (one mode l is a subset of other, i.e., null model (without covariates) is a subs et of full model (with covari ates), the minimum objective function value (OFV) of the best model should be significantly smaller than the alternative model(s) based on the maximum likelihood ratio (MLR) test. The MLR test was applied when the test models fulfilled the full/reduced model definition. A full model can be made equivalent to a reduced model by setting a parameter to a fixed value. The change in OFV between the two nested models is approximately distributed with degree of freedom equal to the number of parame ters that are set to a fixed value in the reduced model. A decrease of 3.84 units in the OFV will be considered statistically significant (p < 0.05) for addition of one para meter during the development of the model. In the case of non-nested models, the different structure models were evaluated based on minimization of the Akaike inform ation criterion (AIC) value. The AIC was defined in terms of: AIC = OFV + 2*p (6-16) where p = total number of parameters in the model (structural + error). Lower the value of AIC better the model. A plot of voriconazole concentration (observed (DV), indi vidual predicted (IPRED) and population predicted (PRED)) versus time for all the individual were drawn, which would give an overall trend of f itted concentrations. It could be seen that for certain individuals the popul ation predictions ar e underpredicted or overpredicted.

PAGE 102

102 The observed and predicted plasma concentra tions would be more randomly distributed across the line of unity for the preferred model. NONMEM obtains IPRED values by the Bayesian POSTHOC option. Using the populati on mean estimate of parameters (prior) and each individual data (likelihood), NONMEM obtains the individual parameter estimates (posterior). From the individual parameter estimates, individual predicted concentrations (IPRED) are obtained. A graph of DV vs. IPRED and DV vs. PRED (goodne ss of fit plots) could be looked for any bias in the predictions. Ideally the points should be uniformly distributed along the line of identity. Normally DV vs. IPRED is much better than DV vs. PRED as PRED contains unexplained variability. Plots of the residuals (RES) vs. PRED and weighted residuals (WRES) vs. PRED (goodness of fit plots) could be looked for any unaccounted heterogeneity in the data. The NONMEM codes were written and the initial estimation of all parameters obtained by WinNonlin was used to run NONMEM. The figures were generated with SPLUS (Statistical Sciences, Version 6.2). Results Probe Recovery The individual probe recovery in each experiment was used to convert the microdialysate concentrations to unbound tissue concentrations and all the values were summarized in Table 6-1. As shown from the tables, the in vivo recovery of voriconazole from retrodialysis methods (RD) were quite similar with 43.3 4.2% (mean SD) (range from 37.6% to 48.1%) for 1 g/ml group and 47.6 4.0% (mean SD) (range from 43.4% to 53.6%) for 2 g/ml group.

PAGE 103

103 Individual Pharmacokinetic Analysis of Total Voriconazole in Plasma after Intravenous Bolus Administration The experimental data of total plas ma voriconazole from 5 and 10 mg/kg intravenous dose were listed in Table 62 and 6-3 and were analyzed by both noncompartmental pharmacokinetic analysis and compartmental pharmacokinetic analysis. Noncompartmental pharm acokinetic analysis After noncompartmental pharmacokinetic anal ysis by WinNonlin 5.2, the results of voriconazole for the total plas ma data after intravenous bol us administration of 5 and 10 mg/kg were listed in Table 6-2 and 6-3. The in itial concentrations (C0) were 3.78 0.29 mg/L (mean SD) and 7.76 0.18 mg/L (mean SD) respectively, which declined with a terminal half-life of 3.95 0.34 hours (mean SD) and 4.92 0.29 hours (mean SD). The areas under the curve AUC were 20.53 2.57 hrmg/L (mean SD) and 56.07 3.82 hrmg/L (mean SD), respectively. The areas under the first moment curve (AUMC ) were 121.90 24.64 hr2mg/L (mean SD) and 417.35 43.75 hr2mg/L (mean SD), respectively. The residence times (MRT ) were 5.89 0.52 hours (mean SD) and 7.43 0.40 hours (mean SD), respecti vely. The volumes of distribution of the central compartment (Vc) were 1.33 0. 1 mL/g (mean SD) and 1.29 0.03 mL/g (mean SD), respectively and Vdss were 1.44 0.09 mL/g (mean SD) and 1.33 0.12 mL/g (mean SD), respectiv ely. The total body clearan ces (CL) were 0.25 0.03 mL/hr/g (mean SD) and 0.18 0.02 mL /hr/g (mean SD), respectively. Compartmental pharmacokinetic analysis The experimental data of total plas ma voriconazole from 5 and 10 mg/kg intravenous dose were plotted in Figure 6-3 and 6-4 and we re analyzed by both oneand

PAGE 104

104 two-compartment model with linear or nonlinear elimination (Michaelis-Menten elimination) by NONMEM. Initially, all four models have applied to fit unbound plasma and muscle voriconazole PK data from 5 and 10 mg/kg intravenous dose, and the OFV and AIC value have been listed in Table 6-4. As shown in this table, the AIC value (344.03) from two-compartment model with non linear elimination was the lowest, which indicated that this model might be the best to fit the total plasma voriconazole PK data in rats. One-compartment model with linear elimination analysis A one-compartment model with linear elimina tion was used to fit the data of total plasma voriconazole from 5 and 10 mg/kg in travenous dose and was able to produce a good curve fit of these data. As shown in Table 64, the OFV is -240.73, and the AIC is 230.73, which is the highest compared to othe r models. The critical parameters obtained in this analysis were listed in Table 6-5. The diagnostic plots for this one-compartment model were also shown in Figur e 6-5, 6-6 and 6-7. As show n in Figure 6-5, the observed and predicted plasma concentrations were randomly distributed ac ross the line of unity. As shown in Figure 6-6, the points of DV, IPRED and PRED were uniformly distributed along the line of identity, and in Figure 6-7, the residuals and weighted residuals plot versus predicted voriconazole concentration fo r this model showed a relatively uniform distribution of residuals ove r the concentration range. Two-compartment model with linear elimination analysis A two-compartment model with linear elimina tion was also used to fit the data of total plasma voriconazole from 5 and 10 mg/kg intravenous dose and was able to produce a good curve fit of these data. As can be seen from Table 6-4, the OFV is -295.77, and the AIC is -277.77. The critical parameters obtained in this analysis were listed in Table

PAGE 105

105 6-6. The diagnostic plots for this two-compar tment model were also shown in Figure 6-8, 6-9 and 6-10. The observed and predicted plasma concentrations were randomly distributed across the line of unity except so me early time point data. The points of DV, IPRED and PRED were uniformly distributed along the line of identity, and the residuals and weighted residuals plot versus predicted voriconazole concentration for this model showed a relatively uniform distribution of residuals over the concentration range. One-compartment model with nonlinear elim ination A one-compartment model with nonlinear elimination was used to fit the data of total plasma voriconazole from 5 and 10 mg/kg intravenous dose and was able to produce a good curve fit of these data. As shown in Table 6-4, the OFV is -249.17, and the AIC is -235.17. The critical parameters obtained in th is analysis were listed in Table 6-7. The diagnostic plots for this one-c ompartment model were also shown in Figure 6-11, 6-12 and 6-13. As shown in Figure 6-11, the plot of voriconazo le concentration (observed (DV), individual predicted (IP RED) and population predicted (PRED)) versus time for the entire individual gave an overall trend of fitted concentrations. The observed and predicted plasma concentrations were randomly distributed across the line of unity. As shown in Figure 6-12, the points of DV, IP RED and PRED were uni formly distributed along the line of identity, and in Figure 6-13, the residuals and weighted residuals plot versus predicted voriconazole concentration fo r this model showed a relatively uniform distribution of residuals ove r the concentration range. Two-compartment model with nonlinear elim ination To find the best model to fit the data of total plasma voriconazole from 5 and 10 mg/kg intravenous dose, a two-compartment model with nonlinear el imination was also applied. This model was able to produce a good curve fit of these data. As shown in

PAGE 106

106 Table 6-4, the OFV is -366. 03, and the AIC is -344.03, which is the lowest compared to other models. The critical parameters obtained in this analysis were listed in Table 6-8. The diagnostic plots for this tw o-compartment model were also shown in Figure 6-14, 6-15 and 6-16. As shown in Figure 6-14, the pl ot of voriconazole c oncentration (observed (DV), individual predicted (IP RED) and population predicted (PRED)) versus time for the entire individual gave an overall trend of fitted concentrations. The observed and predicted plasma concentrations were randomly distributed across the line of unity. As shown in Figure 6-15, the points of DV, IP RED and PRED were uni formly distributed along the line of identity. As can be seen in Figure 6-16, the residuals and weighted residuals plot versus predic ted voriconazole concentration for this model showed a relatively uniform distribution of residua ls over the concentration range. All the diagnostic plots and AIC valu e have indicated that the two-compartmen t model with nonlinear elimination seem to adequately explain the variability of the data and fit the data better than other models. Individual Pharmacokinetic Analysis of Unbound Voriconazole in Muscle after Intravenous Bolus Administration The experimental data of unbound muscle voriconazole concentration from 5 and 10 mg/kg intravenous dose were obtained from rats muscle in vivo microdialysis based on in vivo recovery of voriconazole from retrodi alysis methods (RD) as described previously in this chapter. All the data were listed in Table 6-9, 6-10 and were analyzed by both noncompartmental pharmacokine tic analysis and compartmental pharmacokinetic analysis.

PAGE 107

107 Noncompartmental pharm acokinetic analysis After noncompartmental pharmacokinetic anal ysis, the results of voriconazole for the unbound muscle data after intravenous bolus administra tion of 5 and 10 mg/kg were listed in Table 6-9 and 6-10. The values of terminal half-life were 2.83 0.20 hours (mean SD) and 6.92 2.29 hours (mean SD). The areas under the curve AUC were 7.52 1.02 hrmg/L (mean SD) and 24.52 7.84 hrmg/L (mean SD), respectively. The areas under the first moment curve (AUMC) were 36.72 6.98 hr2mg/L (mean SD) and 267.98 184.84 hr2mg/L (mean SD), respectively. The residence times (MRT) were 4.85 0.31 hours (mean SD) and 10.16 3.02 hours (mean SD), respectively. The total body clearances (CL) were 0.68 0.10 mL/hr/g (mean SD) and 0.43 0.11 mL/hr/g (mean SD), respectively. Compartmental pharmacokinetic analysis Since the two-compartment m odel with nonlinear elimination is the best model to fit the total plasma voriconazole PK data in rats, the experimental data of total plasma and unbound muscle voriconazole PK data fr om 5 and 10 mg/kg intravenous dose were analyzed simultaneously by two-compartm ent model with nonlinear elimination (Michaelis-Menten elimination) by NONM EM. The experimental original unbound muscle voriconazole PK data were plotted in Figure 6-17 and 6-18. Two-compartment model with nonlinear elimination To find the best model to fit the data of unbound plasma and muscle voriconazole from 5 and 10 mg/kg intravenous dose, a two-compartment m odel with nonlinear elimination was applied. This model was also able to produce a good curve fit of these data. The critical parameters obtained in this analysis were listed in Table 6-11. The diagnostic plots for this mode l were also shown in Figure 6-19, 6-20 and 6-21. As shown

PAGE 108

108 in Figure 6-19, the plot of total plasma and unbound muscle voriconazole concentration (observed (DV), indivi dual predicted (IPRED) and populat ion predicted (PRED)) versus time for the entire individual gave an overall trend of fitted concentrations. The observed and predicted plasma concentrations were randomly distributed ac ross the line of unity. As shown in Figure 6-20, the points of DV, IPRED and PRED were uniformly distributed along the line of identity. As can be seen in Figure 6-21, the residuals and weighted residuals plot versus predic ted voriconazole concentration for this model showed a relatively uniform distribution of residuals ove r the concentration range It indicated that the two-compartment model with nonlinear elim ination seem to adequately explain the variability of the unbound plasma and muscle voriconazole PK data. Pharmacokinetic Analysis of Average Total Voriconazole Data in Plasma and Unbound Voriconazole Data in Muscle after Intravenous Bolus Administration The average data of total plasma a nd unbound muscle voriconazole from 5 and 10 mg/kg intravenous dose were listed in Table 6-2, 6-3, 6-9 and 6-10, and were fitted simultaneously by two-compartment model with nonlinear elimination (MichaelisMenten elimination) by NONMEM using the respective parameters of V1= 0.5 L, Q = 0.28 L/hr, Vmax = 0.34 mg/hr, Km = 1.52 mg /L. This model produced a good curve fit of the average plasma and unbound muscle voricon azole concentrations (Figure 6-10 and 611). The data of unbound plasma voriconazole concentration from 5 and 10 mg/kg intravenous dose were obtained from tota l plasma concentration based on a protein binding of 66.2% in rat plasma, which was measured by in vitro microdialysis as described in Chapter 5. The average data of unbound plasma voriconazole were also calculated and plotted in the same figures (Figure 6-10 and 6-11).

PAGE 109

109 Discussion In this study, the pharmacokinetics of voricon azole in male Wistar rats in doses of 5 and 10 mg/kg given intravenous ly was investigated. Plasma concentrations as well as unbound tissue concentrations in thigh muscle were measured. As it is known, antifungal effect relates to the an tifungal agent concentratio ns in plasma and tissue at the target site. The aim was to establish the relationship between plasma concentration and tissue concentration, which may allow us to pred ict the pharmacodynamic ef fect of antifungal agent and help optimize the dosage regimen. Elimination of voriconazole from plasma does not follow simple linear pharmacokinetics in rats which is consistent with the previous report [31]. The voriconazole plasma profiles following intraven ous administration (Fi gure 6-3, 6-4) are convex to some extent indicating the charac teristic of compounds that show capacitylimited elimination. Compared the 5 mg/kg dosing regimen to 10 mg/kg, pharmacokinetic parameters are dependent upon dose. There was superproportional increase in area under the curve was seen with increas ing dose in rat stud ies. There is 2.7fold increase in AUC for a 2fold increase in intravenous do se. It has been reported that voriconazole was eliminated predominantly by metabolism such as the human hepatic cytochrome P450 enzymes of CYP2C19, CYP2C9 and CYP3A4 [31]. The terminal elimination half-lives calculated for voric onazole in was 3.95h for 5mg/kg dosage and 4.92h for 10mg/kg dosage, which is similar to what has been reported in the l iteratures for rats [31, 154]. In Araujo, et als report, the estima ted half-life was found to be 2.4.6 h considering that the pharmacokinetic study was carried out for 8 h. In Roffey, et als report, the voriconazoles plasma PK data we re extracted and concentrations were fitted

PAGE 110

110 to a noncompartment analysis (NCA) model using WinNonlin. The estimated half-life was found to be 4 h considering that the pha rmacokinetic study was carried out for 6h. The elimination of voriconazole was ch aracterized by nonlinear pharmacokinetics in rats. It is likely that saturation of meta bolic clearance is the cau se of the nonlinearity. Compared the fit of linear with nonlinear models, 1-compartment with 2-compartment models, the 2-compartment with nonlinear elimin ation had the best fitting with a lowest AIC value of -344.03. Microdialysis has been reported to be a ve ry useful technique to measure a variety of drugs in different tissues or endogenous compounds in vivo [73, 133, 155]. It is very important to determine the probe recovery when applying microdialysis for a new compound if a good prediction of the true tissu e levels is wanted. The recoveries of voriconazole in rat muscle were determined by retrodialysis method in our studies. Voriconazole exhibits moderate plasma protein binding and is independent of plasma drug concentrations in vitro. The unbound concentrations of voriconazole in the muscle were measured directly and comp ared with the calculated free plasma concentrations using the average plasma prot ein binding of voriconazole in rat of 66.2%. The results showed that muscle had similar concentrations of unbound plasma concentrations. We fitted plasma and tissue concentrations simultaneously. A two-compartment model with nonlinear eliminati on described the data well a nd resulted in a value of f u (0.38) for muscle. The f u can be interpreted as the free fr action of voriconazo le in tissue such as in muscle which might be affected by the subjects covariat es, weight, height, and the plasma albumin concentration [156, 157].

PAGE 111

111 In summary, a two-compartment model w ith non-linear elimination was able to produce a good curve of both the individua l and the average plasma and muscle voriconazole concentrations. The unbound voricon azole concentrations in muscle were almost identical with the calculated unbound plasma concentrations for different dosages indicating the muscle microdial ysis sample may be used for estimating the free plasma levels for voriconazole. The free fr action of voriconazole in tissue (f u ) can also be predicted by PK analysis.

PAGE 112

112 Table 6-1. Recovery data of voriconazole using retr odialysis (RD) method in rats muscle Anima l Recovery (%) 1 g/m l 2 g/m l 139.953.6 241.546.9 337.651.0 446.343.4 548.143.8 646.147.1 Mean (%)43.347.6 SD (%)4.24.0

PAGE 113

113 Table 6-2. Total plasma voriconazole (5 mg/kg) indi vidual noncompartmental pharmacokinetics analysis Time (hr) 515253545556Mea n SDMedian 0.173.463.943.583.523.143.503.520.263.51 0.502.983.353.203.122.753.013.070.213.06 1.002.633.052.822.722.482.542.710.212.67 1.502.392.682.642.592.252.332.480.182.49 2.002.272.542.402.352.092.112.290.172.31 3.002.142.362.222.051.891.792.080.212.10 4.001.862.072.011.831.741.591.850.181.85 5.001.681.731.781.481.461.301.570.191.58 6.001.421.521.521.231.191.011.310.211.32 8.001.031.071.060.840.860.720.930.140.95 Ke [hr-1] 0.160.160.170.190.180.200.180.020.18 t1/2 [hr] 4.264.373.993.673.923.483.950.343.95 C0 [mg/L] 3.734.273.793.743.363.783.780.293.76 AUC[(mg/L)hr] 21.8023.7622.5219.4018.7416.9920.532.5720.60 AUMC [(mg/L)hr2] 140.10150.96137.42105.33110.5987.02121.9024.64124.01 MRT [hr] 6.436.356.105.435.905.125.890.526.00 Vc [mL/g] 1.341.171.321.341.491.321.330.101.33 Vdss [mL/g] 1.471.341.351.401.581.511.440.091.44 CL [mL/hr/g]0.230.210.220.260.270.290.250.030.24 Concentration (mg/L)

PAGE 114

114 Table 6-3. Total plasma voriconazole (10 mg/kg) individual noncompartmental pharmacokinetics analysis Time (hr) 101102103104105106Mea n SDMedian 0.177.167.537.117.617.267.457.350.217.35 0.506.316.806.467.056.336.676.600.296.57 1.005.756.195.936.605.536.226.040.386.06 1.505.346.025.355.625.265.995.600.345.48 2.005.025.375.055.505.105.255.210.195.18 3.004.515.134.765.264.785.004.910.284.89 4.004.305.074.464.874.104.544.560.364.50 5.003.784.564.084.333.944.204.150.284.14 6.003.214.033.613.743.243.943.630.353.68 8.002.392.962.782.812.562.822.720.212.80 Ke [hr-1] 0.150.150.130.140.140.140.140.010.14 t1/2 [hr] 4.564.765.404.814.955.034.920.294.88 C0 [mg/L] 7.637.937.467.917.787.877.760.187.83 AUC[(mg/L)hr] 49.8859.6257.6958.1352.8458.2956.073.8257.91 AUMC [(mg/L)hr2] 342.79438.36465.78422.35392.91441.91417.3543.75430.35 MRT [hr] 6.877.358.077.277.447.587.430.407.39 Vc [mL/g] 1.311.261.341.261.291.271.290.031.28 Vdss [mL/g] 1.381.221.501.181.371.321.330.121.34 CL [mL/hr/g]0.200.170.190.160.180.170.180.010.18 Concentration (mg/L)

PAGE 115

115 Table 6-4. Comparison of objective function value (OFV) and Akaike information criterion (AIC) value from rats total plasma voriconazole PK analysis using different models Models Total number of parameters (p)Objective function value (OFV)Akaike information criterion (AIC) value 1-compartment with linear elimination 5 -240.73 -230.73 1-compartment with nonlinear elimination 7 -249.17 -235.17 2-compartment with linear elimination 9 -295.77 -277.77 2-compartment with nonlinear elimination 11 -366.03 -344.03

PAGE 116

116 Table 6-5. One-compartment model with linear elimination for rat total plasma voriconazole PK analysis Parameter estimatesPopulation estimate (SE%)Between subject variability (BSV) (SE%) CL (L/hr) 0.082 (8.7) 22.4% (37.3) V (L) 0.58 (1.7) 5.1% (31.5) Residual variability Pro p ortional error5.5% ( 6.8 ) Note: % SE: percent standard error of the population parameter estimate. Table 6-6. Two-compartment model with linear elimination for rat total plasma voriconazole PK analysis Parameter estimatesPopulation estimate (SE%)Between subject variability (BSV) (SE%) CL (L/hr)0.070 (6.7)23.5% (28.3) V1 (L) 0.49 (1.7) 6.3% (28.0) Q (L/hr) 0.35 (11.8) 10.0% (-) V2 (L) 0.11 (7.9) 10.0% (-) Residual variability Pro p ortional error 3.9% ( 12.1 ) Note: Dashes indicate that the respective parameters tested during model building were fixed. % SE: percent standard error of the population parameter estimate.

PAGE 117

117 Table 6-7. One-compartment with non-linear elimination model for rat total plasma voriconazole PK analysis Parameter estimatesPopulation estimate (SE%)Between sub j ect variabilit y ( BSV ) ( SE% ) Vmax (mg/hr) 0.86 (24.8) 9.3% (74.7) Km (mg/L) 4.55 (31.0)1.0% (-) V (L) 0.59 (1.5) 5.7% (27.5) Residual variability Pro p ortional error5.5% ( 5.9 ) Note: Dashes indicate that the respective parameters tested during model building were fixed. % SE: percent standard error of the population parameter estimate. Table 6-8. Two-compartment with non-linear elimination model for rat total plasma voriconazole PK analysis Parameter estimatesPopulation estimate (SE%)Between subject variability (BSV) (SE%) V1 (L) 0.50 (1.6) 6.5% (29.8) Q (L/hr) 0.28 (8.6) 10.0% (-) V2 (L) 0.14 (6.5) 11.4% (96.2) Vmax (mg/hr) 0.34 (8.1) 12.0% (42.4) K m (mg/L) 1.52 (17.5) 10.0% (-) Residual variability (SE%) Proportional error 2.9% (15.9) Note: Dashes indicate that the respective parameters tested during model building were fixed. % SE: percent standard error of the population parameter estimate.

PAGE 118

118 Table 6-9. Unbound muscle voriconazole (5 mg/kg) individual noncompartmental pharmacokinetics analysis Time (hr) 515253545556MeanSDMedian 0.331.391.401.701.431.541.481.490.121.45 0.671.161.241.501.171.281.171.250.131.21 1.001.121.191.340.901.271.101.150.151.15 1.331.021.141.130.861.170.891.030.131.07 1.670.901.061.070.801.070.820.950.130.98 2.000.830.941.050.790.980.800.900.110.88 2.330.770.861.000.710.920.710.830.120.82 2.670.740.840.930.670.900.690.800.110.79 3.000.710.800.860.670.880.670.770.090.75 3.330.680.780.820.630.870.650.740.100.73 3.670.650.760.770.580.870.630.710.110.71 4.000.630.740.760.560.830.610.690.100.68 4.330.610.710.700.550.740.580.650.080.65 4.670.590.670.650.490.720.560.610.080.62 5.000.570.650.640.490.690.540.590.080.60 5.330.560.610.600.430.660.540.570.080.58 5.670.550.600.580.420.640.510.550.080.57 6.000.540.570.570.400.620.470.530.080.55 6.330.530.540.550.370.600.450.510.080.54 6.670.510.540.490.350.570.430.480.080.50 7.000.500.510.450.320.550.400.450.090.47 7.330.460.490.420.300.510.380.430.080.44 7.670.410.450.380.270.480.360.390.070.40 8.000.390.410.360.250.440.310.360.070.37 Ke [hr-1] 0.260.270.240.250.220.250.250.020.25 t1/2 [hr] 2.692.612.942.743.172.832.830.202.78 AUC[(mg/L)hr] 7.337.908.226.048.856.777.521.027.62 AUMC [(mg/L)hr2] 36.5938.6138.9326.6447.3432.1836.726.9837.60 MRT [hr] 4.994.894.734.415.354.754.850.314.82 CL [mL/hr/g]0.680.630.610 .830.570.740.680.100.66 Concentration (mg/L)

PAGE 119

119 Table 6-10. Unbound muscle voriconazole ( 10 mg/kg) individual noncompartmental pharmacokinetics analysis Time (hr) 101102103104105106MeanSDMedian 0.332.592.422.372.803.003.252.740.342.69 0.672.422.292.252.522.742.912.520.262.47 1.002.092.001.842.112.222.512.130.232.10 1.332.041.771.572.092.052.301.970.262.04 1.671.821.731.542.012.022.191.880.231.91 2.001.671.691.431.861.962.161.790.261.77 2.331.561.631.371.811.922.151.740.281.72 2.671.481.621.351.761.852.101.690.271.69 3.001.461.471.301.671.821.991.620.261.57 3.331.441.461.281.571.801.951.580.241.52 3.671.401.451.261.521.701.911.540.231.49 4.001.391.371.231.471.681.881.500.241.43 4.331.341.361.221.411.661.861.470.241.39 4.671.301.351.211.351.611.771.430.211.35 5.001.261.321.191.321.551.731.400.201.32 5.331.241.271.181.271.511.711.360.201.27 5.671.221.241.161.231.481.641.330.191.23 6.001.171.221.141.201.461.611.300.191.21 6.331.171.191.121.161.451.601.280.201.18 6.671.151.161.091.131.391.581.250.191.16 7.001.121.131.031.101.321.561.210.201.13 7.331.091.091.001.061.291.531.180.201.09 7.671.061.060.981.011.231.491.140.191.06 8.001.021.030.940.971.191.471.110.201.03 Ke [hr-1] 0.120.090.130.130.120.060.110.030.12 t1/2 [hr] 5.957.575.495.445. 7511.306.922.295.85 AUC[(mg/L)hr] 20.8823.2818.3920.4324.1939.9424.527.8422.08 AUMC [(mg/L)hr2] 186.06254.35156.14162.96210.20 638.19267.98184.84198.13 MRT [hr] 8.9110.938.497.988.6915.9810.163.028.80 CL [mL/hr/g]0.480.420.580 .460.400.250.430.110.44 Concentration (mg/L)

PAGE 120

120 Table 6-11. Rat total plasma and unbound muscle voriconazole PK analysis using twocompartment with non-lin ear elimination model Parameter estimatesPopulation estimate (SE%) B etween sub j ect variabilit y ( BSV ) ( SE% ) V1 (L) 0.50 (1.6) 6.5% (23) Q (L/hr) 0.28 (8.6) 10.0% (-) V2 (L) 0.14 (6.5) 18.7% (70.1) Vmax (mg/hr) 0.34 (8.1) 12.1% (24.6) Km (mg/L) 1.52 (17.5)10.0% (-) fu 0.38 (3.6) 13.0% (41.6) Total p lasma concentration ( SE% ) Unbound muscle concentration (SE%) Proportional error3.1% (21.6)26.5% (13.5) Additive error ( m g /L ) -0 0 3 2 ( ) Residual variabilit y Note: Dashes indicate that the respective parameters tested during model building were fixed. % SE: percent standard error of the population parameter estimate.

PAGE 121

121 Figure 6-1. Cascade pharmacokinetic one-compart ment model with non-linear elimination of voriconazole scheme Figure 6-2. Cascade pharmacokinetic two-compar tment model with non-linear elimination of voriconazole scheme

PAGE 122

122 A Time_hr 02468 Voriconazole conc._mg/L 0.1 1 10 51 52 53 54 55 56 B Time_hr 02468 Voriconazole conc._mg/L 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 51 52 53 54 55 56 Figure 6-3. Dosage of 5 mg/kg i.v. bolus: total voriconazole concentration in rat plasma. A) Linear scale. B) Log scale.

PAGE 123

123 A Time_hr 02468 Voriconazole concentration_mg/L 2 3 4 5 6 7 8 101 102 103 104 105 106 B Time_hr 02468 Voriconazole concentration_mg/L 0.1 1 10 101 102 103 104 105 106 Figure 6-4. Dosage of 10 mg/kg i.v. bolus: total voriconazole con centration in rat plasma. A) Linear scale. B) Log scale.

PAGE 124

124 A B Figure 6-5. The PK plots using one-compartment with linear elimination model for PK analysis of rat total plasma voriconazole data: Plot of observed (), individual predicted (IPRED) (---) and population predicted (PRED) (---) voriconazole concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.

PAGE 125

125 Figure 6-6. Goodness of fit plots for obs erved vs predicte d concentrations ( Observed vs individual predicted, -Observed vs population predicted ) using one-compartment with linear elimination model for PK analysis of rat total plasma voriconazole concentration data. Figure 6-7. Goodness of fit plots for residuals using one-compart ment with linear elimination model for PK analysis of rat total plasma voriconazole concentration data.

PAGE 126

126 A B Figure 6-8. The PK plots using tw o-compartment with linear elimination model for PK analysis of rat total plasma voriconazole data: Plot of observed (), individual predicted (IPRED) (---) and population predicted (PRED) (---) voriconazole concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.

PAGE 127

127 Figure 6-9. Goodness of fit plots for obs erved vs predicte d concentrations ( Observed vs individual predicted, -Observed vs population predicted ) using two -compartment with linear elimination model for PK analysis of rat total plasma voriconazole concentration data. Figure 6-10. Goodness of fit plots for residuals using two-compartment wi th linear elimination model for PK analysis of rat total plasma voriconazole concentration data.

PAGE 128

128 A B Figure 6-11. The PK plots using one-compartment with non-linear elim ination model for PK analysis of rat total plasma voriconazole data: Plot of observed (), individual predicted (IPRED) (---) and population predicted (PRED) (---) voriconazole concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.

PAGE 129

129 Figure 6-12. Goodness of fit plots for obs erved vs predicted concentrations ( Observed vs individual predicted, -Observed vs population predicted ) using one-compartment with non-linear elimin ation model for PK analysis of rat total plasma voriconazole concentration data. Figure 6-13. Goodness of fit plots for residua ls using one-compartment with non-linear elimination model for PK analysis of rat total plasma voriconazole concentration data.

PAGE 130

130 A B Figure 6-14. The PK plots usi ng two-compartment with non-linea r elimination model for PK analysis of rat total plasma voriconazole data: Plot of observed (), individual predicted (IPRED) (---) and population predicted (PRED) (---) voriconazole concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.

PAGE 131

131 Figure 6-15. Goodness of fit plots for obs erved vs predicted concentrations ( Observed vs individual predicted, -Observed vs population predicted ) using two-compartment with non-linear elimin ation model for PK analysis of rat total plasma voriconazole concentration data. Figure 6-16. Goodness of fit plots for residua ls using two-compartment with non-linear elimination model for PK analysis of rat total plasma voriconazole concentration data.

PAGE 132

132 A Time_hr 02468 Voriconazole concentration_mg/L 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 501 502 503 504 505 506 B Time_hr 02468 Voriconazole concentration_mg/L 0.1 1 10 501 502 503 504 505 506 Figure 6-17. Dosage of 5 mg/kg i.v. bolus: unbound voriconazole conc entration in rat muscle. A) Linear scale. B) Log scale.

PAGE 133

133 A Time_hr 02468 Voriconazole concentration_mg/L 0.5 1.0 1.5 2.0 2.5 3.0 3.5 1001 1002 1003 1004 1005 1006 B Time_hr 02468 Voriconazole concentration_mg/L 0.1 1 10 1001 1002 1003 1004 1005 1006 Figure 6-18. Dosage of 10 mg/kg i.v. bolus: unbound voriconazole conc entration in rat muscle. A) Linear scale. B) Log scale.

PAGE 134

134 A B Figure 6-19. The PK plots us ing two-compartment with nonlinear elimination model for analysis of rat total plasma (upper cu rves) and unbound muscle (lower curves) voriconazole data: Plot of observed (), individual predic ted (IPRED) (---) and population predicted (PRED) (---) voriconazole concentration versus time. A) 5 mg/kg dose. B) 10 mg/kg dose.

PAGE 135

135 Figure 6-20. Goodness of fit plots for obs erved vs predicted concentrations ( Observed vs individual predicted, -Observed vs population predicted ) using two-compartment with non-linear elimination model for PK analysis of rat total plasma and unbound muscle voriconazole concentration data. Figure 6-21. Goodness of fit plots for residua ls using two-compartment with non-linear elimination model for PK analysis of rat total plasma and unbound muscle voriconazole concentration data.

PAGE 136

136 A Time (hr) 02468 Voriconazole Concentration (mg/L) 0 1 2 3 4 5 Cptotal Predicted Cptotal Cpfree Cmunbound Predicted Cmunbound B Time (hr) 02468 Voriconazole Concentration (mg/L) 0.1 1 10 Cptotal Predicted Cptotal Calculated Cpfree Cmunbound Predicted Cmunbound Figure 6-22. Dosage of 5 mg/kg i.v. bolus in rat: average plasma and muscle PK data analysis using two-compartment with nonlinear elimination model. A) Linear scale. B) Log scale.

PAGE 137

137 A Time (hr) 02468 Voriconazole Concentration (mg/L) 0 2 4 6 8 10 Cptotal Predicted Cptotal Calculated Cpfree Cmunbound Predicted Cmunbound B Time (hr) 02468 Voriconazole Concentration (mg/L) 0.1 1 10 Cptotal Predicted Cptotal Calculated Cpfree Cmunbound Predicted Cmunbound Figure 6-23. Dosage of 10 mg/kg i.v. bolus in rat: average plasma and muscle PK data analysis using two-compartment with nonlinear elimination model. A) Linear scale. B) Log scale.

PAGE 138

138 CHAPTER 7 CONCLUSIONS The pharm acokinetic and pharmacodynamic charact eristics of voriconazole have been reported in many publications. Howe ver, the correlation between in vitro activity and pharmacokinetic/pharmacodynamic parameters of vor iconazole with its c linical outcome in fungal infections need to be furt her studied. There is need for mathematical models of this drug to guide its optimal use. We did the time-kill and postantifungaleff ect (PAFE) experiments for voriconazole against Candida albicans Candida glabrata, and Candida parapsilosis isolates. Moreover, a high-performance liquid chromatography (HPLC) assay was developed to validate these experiments. Our findings demonstrated that vor iconazole exerts prolong ed fungistatic activity against C. albicans, C. glabrata and C. parapsilosis but no PAFE at concentrations achievable in human sera with routine dosing. These finding s are potentially relevant clinically with other antifungal agents, to which certain Candida isolates exhibit diminished susceptibility or develop resistance. HPLC confirmed that experiment s were conducted at the desired steady-state voriconazole concentrations. To our knowledge, this is the first study to verify standard time-kill and PAFE methodologies by direc tly measuring drug concentrati ons. These HPLC methods were essential to the design of dynamic in vitro models to assess the pharmacodynamics of voriconazole and other agents prior to the achievement of steady-state conditions. We developed a pharmacokinetic/pharmacodynami c (PK/PD) mathematical model that fits voriconazole timekill data against Candida isolates in vitro and used the model to simulate the expected kill curves for t ypical intravenous and oral dos ing regimens. A series of Emax mathematical models were used to fit timekill data for two isolates each of Candida albicans Candida glabrata and Candida parapsilosis PK parameters extracted from human data sets

PAGE 139

139 were used in the model to simulate kill curves for each isolate. Timekill data were best fit by using an adapted sigmoidal Emax model that corrected for delays in candidal growth and the onset of voriconazole activity, sa turation of the number of Candida and the steepness of the concentrationresponse curve. The rates of maximal killing by voriconazole ( kmax) were highly correlated with the growth rates (ks) of the isolates (Pearsons correlation coefficient = 0.9861). Simulations using PK parameters derived from th e human data sets showed fungistatic effects against each of the isolates. The developed ma thematical PK/PD model linked pharmacokinetic and pharmacodynamic of voriconazole to provide the basic understanding for defining optimal antifungal dosing regimens and pred ict antifungal treatment efficacy. We established a dynamic system to mimic the in vivo conditions and the PK/PD relationship of voriconazole agai nst Candida was accurately modeled. Modeling approaches that utilized human PK data were adapted to defi ne the optimal use of voriconazole and other antifungal agents. Microdialysis provided a simple, clean method for determini ng the free drug concentration in vitro and in vivo In order to confirm that the perfor mance of voriconazole microdialysis was feasible, we conducted the voric onazole protei n binding with in vitro microdialysis method. The pharmacokinetics of voriconazole in vivo and modeling were furthe r investigated. Plasma concentrations as well as unbound tissue concentrations in rat thigh muscle were measured. A two-compartment model with non-li near elimination was used to fit both the individual and the average plasma and muscle vor iconazole concentrations and able to produce good curves. The results from animal study indicated that it is possible to use the free muscle concentrations as a surrogate marker for the free concentrations in plasma. The free fraction of voriconazole in tissue (f u ) was predicted by PK analysis.

PAGE 140

140 In the past, PK/PD approaches mainly based on the comparisons of pharmacokinetics of total plasma concentrations and in vitro MIC. Our PK/PD approach is a big improvement over the previously used. The devel oped PK/PD modeling approaches enables us to make rational antifungal agent dosing decisions by predicting the effect of vari ous dosing regimens, taking into account the clinically e ffective free concentratio ns at the target site against candidiasis.

PAGE 141

141 LIST OF REFERENCES 1. Gerzenshtein L, Patel SM, Scarsi KK, Po stelnick MJ, Flaherty JP. Breakthrough Candida infections in patients receiv ing voriconazole. The Annals of pharmacotherapy. 2005 JulAug;39(7-8):1342-5. 2. Fratti RA, Belanger PH, Sanati H, Gha nnoum MA. The effect of the new triazole, voriconazole (UK-109,496), on the interactions of Candida albicans and Candida krusei with endothelial cells. Journal of chemotherapy (Florence, Italy). 1998 Feb;10(1):7-16. 3. Fukuoka T, Johnston DA, Winslow CA, de Groot MJ, Burt C, Hitchcock CA, et al. Genetic basis for differential activities of fl uconazole and voriconazole against Candida krusei. Antimicrobial agents and chemotherapy. 2003 Apr;47(4):1213-9. 4. Pfizer Inc. Label: voriconazole for injection, tablets, oral suspension: LAB-0271-12. 2005 Mar. 5. Boucher HW, Groll AH, Chiou CC, Wals h TJ. Newer systemic antifungal agents : pharmacokinetics, safety and efficacy. Drugs. 2004;64(18):1997-2020. 6. Sheehan DJ, Hitchcock CA, Sibley CM. Cu rrent and emerging azole antifungal agents. Clin Microbiol Rev. 1999 Jan;12(1):40-79. 7. Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin In fect Dis. 1998 Jan;26(1):1-10. 8. Theuretzbacher U, Ihle F, Derendorf H. Pharmacokinetic/pharmacodynamic profile of voriconazole. Clin Pharmacokinet. 2006;45(7):649-63. 9. Pfizer Inc. Label: voriconazole for injection, tablets, oral suspension: LAB-0271-12. 2005 Mar. 10. Patterson TF. New agents for treatment of i nvasive aspergillosis. Clin Infect Dis. 2002 Aug 15;35(4):367-9. 11. Sambatakou H, Dupont B, L ode H, Denning DW. Voriconazole treatment for subacute invasive and chronic pulmonary aspergillosi s. The American jour nal of medicine. 2006 Jun;119(6):527 e17-24. 12. Camuset J, Nunes H, Dombret MC, Bergeron A, Henno P, Philippe B, et al. Treatment of chronic pulmonary aspergillo sis by voriconazole in noni mmunocompromised patients. Chest. 2007 May;131(5):1435-41. 13. Scott LJ, Simpson D. Voriconazole : a review of its use in the management of invasive fungal infections. Drugs. 2007;67(2):269-98. 14. Pearson MM, Rogers PD, Cleary JD, Ch apman SW. Voriconazole: a new triazole antifungal agent. The Annals of pha rmacotherapy. 2003 Mar;37(3):420-32.

PAGE 142

142 15. Espinel-Ingroff A, Boyle K, Sheehan DJ. In vitro antifungal activ ities of voriconazole and reference agents as determined by NCCLS methods: review of the literature. Mycopathologia. 2001;150(3):101-15. 16. Manavathu EK, Ramesh MS, Baskaran I, Ganesan LT, Chandrasek ar PH. A comparative study of the post-antifunga l effect (PAFE) of amphotericin B, triazoles and echinocandins on Aspergillus fumigatus and Candida al bicans. The Journal of antimicrobial chemotherapy. 2004 Feb;53(2):386-9. 17. Lyman CA, Walsh TJ. Systemically admini stered antifungal agents A review of their clinical pharmacology and therapeutic applications. Drugs. 1992 Jul;44(1):9-35. 18. Odds FC, Cheesman SL, Abbott AB. Antif ungal effects of fluconazole (UK 49858), a new triazole antifungal, in vitro. J An timicrob Chemother. 1986 Oct;18(4):473-8. 19. Como JA, Dismukes WE. Oral azole drugs as systemic antif ungal therapy. N Engl J Med. 1994 Jan 27;330(4):263-72. 20. Sanati H, Belanger P, Fratti R, Gha nnoum M. A new triazo le, voriconazole (UK109,496), blocks sterol bios ynthesis in Candida albi cans and Candida krusei. Antimicrobial agents and chem otherapy. 1997 Nov;41(11):2492-6. 21. Belanger P, Nast CC, Frat ti R, Sanati H, Ghannoum M. Voriconazole (UK-109,496) inhibits the growth and alte rs the morphology of fluconazole-susceptible and -resistant Candida species. Antimicrobial agents and chemotherapy. 1997 Aug;41(8):1840-2. 22. Li Y, Nguyen MH, Derendorf H, Cheng S, Clancy CJ. Measurement of voriconazole activity against Candida albicans, C. glabrata, and C. parapsilosis isolates using time-kill methods validated by high-performance liquid chromatography. Antimicrobial agents and chemotherapy. 2007 Aug;51(8):2985-7. 23. Espinel-Ingroff A. Mechanisms of resistan ce to antifungal agents: Yeasts and filamentous fungi. Rev Iberoam Micol. 2008 Jun;25(2):101-6. 24. Peman J, Salavert M, Canton E, Jarque I, Ro ma E, Zaragoza R, et al. Voriconazole in the management of nosocomial invasive fungal in fections. Therapeutics and clinical risk management. 2006 Jun;2(2):129-58. 25. Quindos G, Carrillo-Munoz AJ, Eraso E, Ca nton E, Peman J. In vitro antifungal activity of voriconazole: New data after the first years of clinical experience. Rev Iberoam Micol. 2007 Sep 30;24(3):198-208. 26. Espinel-Ingroff A, Arthington-Skaggs B, Iqbal N, Ellis D, Pfaller MA, Messer S, et al. Multicenter evaluation of a new disk agar di ffusion method for sus ceptibility testing of filamentous fungi with voriconazole, posaconazole, itraconazole, amphotericin B, and caspofungin. Journal of clinical microbiology. 2007 Jun;45(6):1811-20.

PAGE 143

143 27. Borgers M. Mechanism of action of antif ungal drugs, with special reference to the imidazole derivatives. Rev Infect Dis. 1980 Jul-Aug;2(4):520-34. 28. Muller FM, Weig M, Peter J, Walsh TJ. Azole cross-resistance to ketoconazole, fluconazole, itraconazole and voriconazole in clinical Candida albicans isolates from HIV-infected children with oropharyngeal candidosis. J Antimicrob Chemother. 2000 Aug;46(2):338-40. 29. Cadrobbi J, Hecq J, Lebrun C, Vanbeckbe rgen D, Jamart J, Galanti L. Long-term stability of voriconazole 4 mg/mL in dextrose 5% polyvinyl chloride bags at 4C. The European Journal of Hospital Phar macy Science. 2006;12 (3 ):57 9. 30. Langman LJ, Boakye-Agyeman F. Measurem ent of voriconazole in serum and plasma. Clin Biochem. 2007 Dec;40(18):1378-85. 31. Roffey SJ, Cole S, Comby P, Gibson D, Jezequel SG, Nedderman AN, et al. The disposition of voriconazole in mouse, ra t, rabbit, guinea pig, dog, and human. Drug Metab Dispos. 2003 Jun;31(6):731-41. 32. Andes D, Marchillo K, Stamstad T, Conklin R. In vivo pharmacokinetics and pharmacodynamics of a new triazole, voricona zole, in a murine candidiasis model. Antimicrobial agents and chem otherapy. 2003 Oct;47(10):3165-9. 33. Sugar AM, Liu XP. Effect of grapefruit juic e on serum voriconazole concentrations in the mouse. Med Mycol. 2000 Jun;38(3):209-12. 34. Saunte DM, Simmel F, Frimodt-Moller N, Stol le LB, Svejgaard EL, H aedersdal M, et al. In vivo efficacy and pharmacokinetics of voriconazole in an animal model of dermatophytosis. Antimicrobial agents a nd chemotherapy. 2007 Sep;51(9):3317-21. 35. Damle B, LaBadie R, Crownover P, Glue P. Pharmacokinetic intera ctions of efavirenz and voriconazole in healthy vol unteers. British jour nal of clinical pharmacology. 2008 Apr;65(4):523-30. 36. Quintard H, Papy E, Massias L, Lasock i S, Arnaud P, Desmonts JM, et al. The pharmacokinetic profile of voriconazole during continuous high-volume venovenous hemofiltration in a critically ill patient. Therapeutic drug monitoring. 2008 Feb;30(1):117-9. 37. Jeu L, Piacenti FJ, Lyakhovetskiy AG Fung HB. Voriconazole. Clin Ther. 2003 May;25(5):1321-81. 38. Sabo JA, Abdel-Rahman SM. Voricon azole: a new triazole antifungal. Ann Pharmacother. 2000 Sep;34(9):1032-43. 39. Purkins L, Wood N, Ghahramani P, Greenhalgh K, Allen MJ, Kleinermans D. Pharmacokinetics and safety of voriconazo le following intravenousto oral-dose escalation regimens. Antimicrobial agents and chemotherapy. 2002 Aug;46(8):2546-53.

PAGE 144

144 40. Lazarus HM, Blumer JL, Yanovich S, Schlamm H, Romero A. Safety and pharmacokinetics of oral voriconazole in patie nts at risk of fungal infection: a dose escalation study. Journal of clinical pharmacology. 2002 Apr;42(4):395-402. 41. Arikan S, Lozano-Chiu M, Paetznick V, Nangia S, Rex JH. Microd ilution susceptibility testing of amphotericin B, itraconazole, and voriconazole against clinical isolates of Aspergillus and Fusarium species. J C lin Microbiol. 1999 Dec;37(12):3946-51. 42. Espinel-Ingroff A. In vitro activity of the new triazole vor iconazole (UK-109,496) against opportunistic filamentous and dimorphic fungi and common and emerging yeast pathogens. J Clin Microbio l. 1998 Jan;36(1):198-202. 43. McGinnis MR, Pasarell L, Sutton DA, Fothergill AW, Cooper CR, Jr., Rinaldi MG. In vitro activity of voriconazole against selected fungi. Med Mycol. 1998 Aug;36(4):23942. 44. Clancy CJ, Nguyen MH. In vitro efficacy a nd fungicidal activity of voriconazole against Aspergillus and Fusarium species. Eur J C lin Microbiol Infect Dis. 1998 Aug;17(8):5735. 45. Cordeiro RA, Brilhante RS, Rocha MF, F echine MA, Costa AK, Camargo ZP, et al. In vitro activities of caspofungin, amphotericin B and azoles against Coccidioides posadasii strains from Northeast, Brazil. My copathologia. 2006 Jan;161(1):21-6. 46. Klepser ME, Malone D, Le wis RE, Ernst EJ, Pfaller MA. Evaluation of voriconazole pharmacodynamics using time-kill methodology. Antimicrobial agents and chemotherapy. 2000 Jul;44(7):1917-20. 47. Lignell A, Lowdin E, Cars O, Sjolin J. Characterization of the inhibitory effect of voriconazole on the fungicidal activity of amphotericin B agains t Candida albicans in an in vitro kinetic model. The Journal of antimicrobial chemotherapy. 2008 Apr 12. 48. Lozano-Chiu M, Arikan S, Paetznick VL Anaissie EJ, Rex JH. Optimizing voriconazole susceptibility testing of Candida: effects of incubation time, endpoint rule, species of Candida, and level of fluconazole suscepti bility. J Clin Micr obiol. 1999 Sep;37(9):27559. 49. Nguyen MH, Yu CY. Voricona zole against fluconazole-susce ptible and resistant candida isolates: in-vitro efficacy compared with that of itraconazole and ketoconazole. J Antimicrob Chemother. 1998 Aug;42(2):253-6. 50. Johnson EM, Szekely A, Warnock DW. In-v itro activity of voric onazole, itraconazole and amphotericin B against filamentous fungi. J Antimicrob Chemother. 1998 Dec;42(6):741-5. 51. Dannaoui E, Meletiadis J, Mouton JW, Meis JF, Verweij PE. In vitro susceptibilities of zygomycetes to conventional and new antifungals. J Antimicrob Chemother. 2003 Jan;51(1):45-52.

PAGE 145

145 52. Varkey JB, Perfect JR. Rare and emergi ng fungal pulmonary infections. Seminars in respiratory and critical care medicine. 2008 Apr;29(2):121-31. 53. Alanio A, Brethon B, Feuilhade de Chauvin M, de Kerviler E, Lebl anc T, Lacroix C, et al. Invasive pulmonary infection due to Trichoderma longibrachiatum mimicking invasive Aspergillosis in a neutropenic pa tient successfully treat ed with voriconazole combined with caspofungin. Clin In fect Dis. 2008 May 15;46(10):e116-8. 54. Martin MV, Yates J, Hitchcock CA. Co mparison of voriconazole (UK-109,496) and itraconazole in prevention and treatment of Aspergillus fumigatus endocarditis in guinea pigs. Antimicrobial agents and ch emotherapy. 1997 Jan;41(1):13-6. 55. Kirkpatrick WR, McAtee RK, Fothergill AW, Rinaldi MG, Patterson TF. Efficacy of voriconazole in a guinea pig model of dissemina ted invasive aspergillosis. Antimicrobial agents and chemotherapy. 2000 Oct;44(10):2865-8. 56. Capilla J, Guarro J. Correlation between in vitro susceptibility of Scedosporium apiospermum to voriconazole and in vivo out come of scedosporiosis in guinea pigs. Antimicrobial agents and chem otherapy. 2004 Oct;48(10):4009-11. 57. Lionakis MS, Lewis RE, May GS, Wiederhol d NP, Albert ND, Halder G, et al. Tolldeficient Drosophila flies as a fast, high-throughput model for the study of antifungal drug efficacy against invasive aspergillosis and Aspergillus virulence. J Infect Dis. 2005 Apr 1;191(7):1188-95. 58. Weerasuriya N, Snape J. Oesophageal candi diasis in elderly patie nts: risk factors, prevention and management. Drugs & aging. 2008;25(2):119-30. 59. Heinz WJ, Einsele H. Caspofungin for trea tment of invasive as pergillus infections. Mycoses. 2008;51 Suppl 1:47-57. 60. Proia LA, Tenorio AR. Successful use of voriconazole for treatme nt of Coccidioides meningitis. Antimicrobial agents an d chemotherapy. 2004 Jun;48(6):2341. 61. Stanzani M, Vianelli N, Bandi ni G, Paolini S, Arpinati M, Bonifazi F, et al. Successful treatment of disseminated Fu sariosis after allogeneic hematopoietic stem cell transplantation with the combination of voric onazole and liposomal amphotericin B. The Journal of infection. 2006 Dec;53(6):e243-6. 62. Nesky MA, McDougal EC, Peacock Jr JE Pseudallescheria boydii brain abscess successfully treated with voriconazole and su rgical drainage: case report and literature review of central nervous system pseudallescherias is. Clin Infect Dis. 2000 Sep;31(3):673-7. 63. Kappe R. Antifungal activity of the new azole UK-109, 496 (voriconazole). Mycoses. 1999;42 Suppl 2:83-6.

PAGE 146

146 64. Ally R, Schurmann D, Kreisel W, Carosi G, Aguirrebengoa K, Dupont B, et al. A randomized, double-blind, double-dummy, multicenter trial of voriconazole and fluconazole in the treatment of esophageal candidiasis in immunocompromised patients. Clin Infect Dis. 2001 Nov 1;33(9):1447-54. 65. Kullberg BJ, Sobel JD, Ruhnke M, Pappas PG Viscoli C, Rex JH, et al. Voriconazole versus a regimen of amphotericin B followed by fluconazole for candidaemia in nonneutropenic patients: a randomised non-inferiority trial. Lancet. 2005 Oct 2228;366(9495):1435-42. 66. Hwang YK, Joo NH, Tee GY, Ta n P, Tan L, Girija R. Clinical activity of the new triazole drug voriconazole (UK 109, 496) against disseminated hepatosplenic aspergillosis in a patient w ith relapsed leukemia. Haem atologia (Budap). 2001;31(1):7380. 67. Herbrecht R. Voriconazole: therapeutic review of a new azole antifungal. Expert review of anti-infective thera py. 2004 Aug;2(4):485-97. 68. Singh N, Limaye AP, Forrest G, Safdar N, Munoz P, Pursel l K, et al. Combination of voriconazole and caspofungin as primary therapy for invasive aspergillosis in solid organ transplant recipients: a pros pective, multicenter, observati onal study. Transplantation. 2006 Feb 15;81(3):320-6. 69. de la Pena A, Liu P, Derendorf H. Micr odialysis in peripheral tissues. Adv Drug Deliv Rev. 2000 Dec 15;45(2-3):189-216. 70. Yokoyama M, Suzuki E, Sato T, Maruta S, Watanabe S, Miyaoka H. Amygdalic levels of dopamine and serotonin rise upon exposure to conditioned fear stress without elevation of glutamate. Neurosci Le tt. 2005 Apr 29;379(1):37-41. 71. Humpel C, Ebendal T, Ol son L. Microdialysis: a way to study in vivo release of neurotrophic bioactivity: a critical summary. J Mol Med. 1996 Sep;74(9):523-6. 72. Lee SC, Fung CP, Huang JS, Tsai CJ, Chen KS, Chen HY, et al. Clinical correlates of antifungal macrodilution suscepti bility test results for non-AI DS patients with severe Candida infections treated with fluconazole Antimicrobial agents and chemotherapy. 2000 Oct;44(10):2715-8. 73. Liu P, Fuhrherr R, Webb AI, Oberma nn B, Derendorf H. Tissue penetration of cefpodoxime into the skeletal muscle and lung in rats. Eur J Pharm Sci. 2005 JulAug;25(4-5):439-44. 74. Liu P, Muller M, Grant M, Obermann B, Derendorf H. Tissue penetration of cefpodoxime and cefixime in healthy subjects. Journal of clinical pharmacology. 2005 May;45(5):564-9.

PAGE 147

147 75. Traunmuller F, Zeitlinger M, Zeleny P, Muller M, Joukhadar C. Pharmacokinetics of singleand multiple-dose oral clarithromycin in soft tissues determined by microdialysis. Antimicrobial agents and chem otherapy. 2007 Sep;51(9):3185-9. 76. Hartings JA, Gugliotta M, Gilman C, St rong AJ, Tortella FC, Bullock MR. Repetitive cortical spreading depol arizations in a case of severe br ain trauma. Neurological research. 2008 Jun 3. 77. Hutschala D, Skhirtladze K, Kinstner C, Mayer-Helm B, Muller M, Wolner E, et al. In vivo microdialysis to measure antibiotic pe netration into soft tissue during cardiac surgery. The Annals of thoraci c surgery. 2007 Nov;84(5):1605-10. 78. Tyvold SS, Solligard E, Lyng O, Steinsha mn SL, Gunnes S, Aadahl P. Continuous monitoring of the bronchial ep ithelial lining fluid by microdial ysis. Respiratory research. 2007;8:78. 79. Muller M, Brunner M, Schmid R, Mader RM, Bockenheimer J, Steger GG, et al. Interstitial methotrexate kinetics in primar y breast cancer lesions. Cancer research. 1998 Jul 15;58(14):2982-5. 80. Hansen DK, Davies MI, Lunte SM, Lunte CE. Pharmacokinetic and metabolism studies using microdialysis sampling. J Pharm Sci. 1999 Jan;88(1):14-27. 81. de Lange, Elizabeth C. M., de Boer, A. G., Breimer DD. Methodological issues in microdialysis sampling for pharmacokinetic st udies. Advanced Drug Delivery Reviews. 2000 2000/12/15;45(2-3):125-48. 82. Parent M, Bush D, Rauw G, Master S, Vaccarino F, Baker G. Analysis of Amino Acids and Catecholamines, 5-Hydroxytryptamine and Their Metabolites in Brain Areas in the Rat Using in Vivo Microdialysi s. Methods. 2001 2001/ 1;23(1):11-20. 83. Day JC, Kornecook TJ, Quirion R. Applicati on of in vivo microdia lysis to the study of cholinergic systems. Methods. 2001 Jan;23(1):21-39. 84. Hows ME, Lacroix L, He idbreder C, Organ AJ, Shah AJ. High-performance liquid chromatography/tandem mass spectrometric a ssay for the simultaneous measurement of dopamine, norepinephrine, 5-hydroxytryptamine and cocaine in biological samples. J Neurosci Methods. 2004 Sep 30;138(1-2):123-32. 85. Baseski HM, Watson CJ, Cellar NA, Shackman JG, Kennedy RT. Capillary liquid chromatography with MS3 for the determination of enkephalins in microdialysis samples from the striatum of anesthetized a nd freely-moving rats. J Mass Spectrom. 2005 Feb;40(2):146-53. 86. Bowser MT, Kennedy RT. In vivo monitoring of amine neurotransmitters using microdialysis with on-line capillary electrophoresis. Electrophoresis. 2001 Oct;22(17):3668-76.

PAGE 148

148 87. Huynh BH, Fogarty BA, Martin RS, Lunt e SM. On-line coupling of microdialysis sampling with microchip-based capillary electrophoresis. Anal Chem. 2004 Nov 1;76(21):6440-7. 88. Kanamori K, Ross BD. Suppression of glia l glutamine release to the extracellular fluid studied in vivo by NMR and mi crodialysis in hyperammonemi c rat brain. J Neurochem. 2005 Jul;94(1):74-85. 89. Lucas LH, Wilson SF, Lunte CE, Larive CK. Concentration profiling in rat tissue by high-resolution magic-an gle spinning NMR spectroscopy: in vestigation of a model drug. Anal Chem. 2005 May 1;77(9):2978-84. 90. Mayer BX, Namiranian K, Dehghanyar P, St roh R, Mascher H, Muller M. Comparison of UV and tandem mass spectrometric detec tion for the high-performance liquid chromatographic determination of diclofenac in microdialysis samples. J Pharm Biomed Anal. 2003 Nov 24;33(4):745-54. 91. Kanamori K, Kondrat RW, Ross BD. 13C en richment of extracellular neurotransmitter glutamate in rat brain --combined mass spectrometry and NMR studies of neurotransmitter turnover and uptake into glia in vivo. Ce ll Mol Biol (Noisy-le-grand). 2003 Jul;49(5):819-36. 92. Lindon JC. HPLC-NMR-MS: past, presen t and future. Drug Discov Today. 2003 Nov 15;8(22):1021-2. 93. Lindon JC, Nicholson JK, Wilson ID. Dir ectly coupled HPLC-N MR and HPLC-NMRMS in pharmaceutical research and developm ent. J Chromatogr B Biomed Sci Appl. 2000 Oct 1;748(1):233-58. 94. Lommen A, Godejohann M, Ve nema DP, Hollman PC, Spraul M. Application of directly coupled HPLC-NMR-MS to the identification and confirmation of que rcetin glycosides and phloretin glycosides in apple peel. Anal Chem. 2000 Apr 15;72(8):1793-7. 95. Li Y, Peris J, Zhong L, Derendorf H. Microdialysis as a tool in local pharmacodynamics. The AAPS journal. 2006;8(2):E222-35. 96. Weikop P, Egestad B, Kehr J. Applicati on of triple-probe microdialysis for fast pharmacokinetic/pharmacodynamic evaluati on of dopamimetic activity of drug candidates in the rat brain. Journal of neuroscience methods. 2004 Dec 30;140(1-2):5965. 97. Feng MR, Turluck D, Burleigh J, Lister R, Fan C, Middlebr ook A, et al. Brain microdialysis and PK/PD correlation of pregab alin in rats. European journal of drug metabolism and pharmacokinetics 2001 Jan-Jun;26(1-2):123-8. 98. Au-Yeung SC, Riggs KW, Gruber N, Rurak DW. The use of microdialysis for the study of drug kinetics: central ner vous system pharmacokinetics of diphenhydramine in fetal,

PAGE 149

149 newborn, and adult sheep. Drug metabolism and disposition: the biological fate of chemicals. 2007 Aug;35(8):1285-91. 99. Schuck VJ, Rinas I, Derendorf H. In vitro mi crodialysis sampling of docetaxel. Journal of pharmaceutical and biomedical analysis. 2004 Nov 19;36(4):807-13. 100. Derendorf H, Lesko LJ, Chaikin P, Colburn WA, L ee P, Miller R, et al. Pharmacokinetic/pharmacodynamic modeling in drug research and development. Journal of clinical pharmacology. 2000 Dec;40(12 Pt 2):1399-418. 101. Meibohm B, Derendorf H. Basic con cepts of pharmacokinetic/pharmacodynamic (PK/PD) modelling. International journal of clinical pharmacology and therapeutics. 1997 Oct;35(10):401-13. 102. Li Y, Nguyen MH, Cheng S, Schmid t S, Zhong L, Derendorf H, et al. A pharmacokinetic/pharmacodynamic mathematical model accurately describes the activity of voriconazole against Candida spp. in vitr o. International jour nal of antimicrobial agents. 2008 Apr;31(4):369-74. 103. Xu J, Winkler J, Derendorf H. A pharm acokinetic/pharmacodynamic approach to predict total prednisolone concentrations in huma n plasma. Journal of pharmacokinetics and pharmacodynamics. 2007 Jun;34(3):355-72. 104. Derendorf H, Hochhaus G, Krishnaswam i S, Meibohm B, Mollmann H. Optimized therapeutic ratio of inhaled corticosteroids using retromet abolism. Die Pharmazie. 2000 Mar;55(3):223-7. 105. Derendorf H, Meibohm B. Modeling of pharmacokinetic/pharmacodynamic (PK/PD) relationships: concepts and perspectives. Pharmaceutical research. 1999 Feb;16(2):17685. 106. Walsh TJ, Karlsson MO, Dris coll T, Arguedas AG, Adamson P, Saez-Llorens X, et al. Pharmacokinetics and safety of intravenous voriconazole in childre n after singleor multiple-dose administration. Antimicrobial agents and chemotherapy. 2004 Jun;48(6):2166-72. 107. Louie A, Liu QF, Drusano GL, Liu W, Maye rs M, Anaissie E, et al. Pharmacokinetic studies of fluconazole in rabbi ts characterizing doses which ac hieve peak levels in serum and area under the concentration-time curv e values which mimic those of high-dose fluconazole in humans. Antimicrobial agen ts and chemotherapy. 1998 Jun;42(6):1512-4. 108. Ohtani Y, Kotegawa T, Tsutsumi K, Mo rimoto T, Hirose Y, Nakano S. Effect of fluconazole on the pharmacokinetics and pharmacodynamics of oral and rectal bromazepam: an application of electroence phalography as the pharmacodynamic method. Journal of clinical pharmaco logy. 2002 Feb;42(2):183-91.

PAGE 150

150 109. Louie A, Drusano GL, Banerjee P, Liu QF, Liu W, Kaw P, et al. Pharmacodynamics of fluconazole in a murine model of system ic candidiasis. Antimicrobial agents and chemotherapy. 1998 May;42(5):1105-9. 110. Katayama M, Igarashi H, Tani K, Nezu Y, Harada Y, Yogo T, et al. Effect of multiple oral dosing of fluconazole on the pharmacokinetics of cycl osporine in hea lthy beagles. The Journal of veterinary medi cal science / the Japanese Soci ety of Veterinary Science. 2008 Jan;70(1):85-8. 111. Carrasco-Portugal Mdel C, Flores-Murrieta FJ. Gender differences in the oral pharmacokinetics of fluconazole. Clinical drug investigation. 2007;27(12):851-5. 112. Saari TI, Laine K, Neuvonen M, Neuvonen PJ Olkkola KT. Effect of voriconazole and fluconazole on the pharmacokinetics of intravenous fentanyl. European journal of clinical pharmacology. 2008 Jan;64(1):25-30. 113. Andes D, Marchillo K, Stamstad T, Conklin R. In vivo pharmacodynamics of a new triazole, ravuconazole, in a murine candidiasis model. Antimicrobial agents and chemotherapy. 2003 Apr;47(4):1193-9. 114. Andes D, Marchillo K, Conklin R, Kr ishna G, Ezzet F, Cacciapuoti A, et al. Pharmacodynamics of a new triazole, posaconazole, in a murine model of disseminated candidiasis. Antimicrobial agents an d chemotherapy. 2004 Jan;48(1):137-42. 115. Greer ND. Posaconazole (Noxafil): a new tr iazole antifungal agent. Proc (Bayl Univ Med Cent). 2007 Apr;20(2):188-96. 116. Courtney R, Pai S, Laughlin M, Lim J, Batra V. Pharmacokinetics, safety, and tolerability of oral posaconazole administered in si ngle and multiple doses in healthy adults. Antimicrob Agents Chemother. 2003 Sep;47(9):2788-95. 117. Janssen-Pharmaceutica. Product info rmation: SPORANOX (ITRACONAZOLE) ORAL SUSPENSION: Hepatotoxi city Labeling 2002 June. 118. Pfizer. Product information: DIFLUCAN (Fluconazole): LAB-0099 .1. 2004 August. 119. Clancy CJ, Huang H, Cheng S, Derendorf H, Nguyen MH. Characterizing the effects of caspofungin on Candida albicans, Candida parapsilosis, and Ca ndida glabrata isolates by simultaneous time-kill and postantifungal-effect experiments. Antimicrobial agents and chemotherapy. 2006 Jul;50(7):2569-72. 120. Gage R, Stopher DA. A rapi d HPLC assay for voriconazole in human plasma. J Pharm Biomed Anal. 1998 Sep;17(8):1449-53. 121. MacCallum DM, Whyte JA, Odds FC. Efficacy of caspofungin and voriconazole combinations in experimental aspergi llosis. Antimicrob Agents Chemother. 2005 Sep;49(9):3697-701.

PAGE 151

151 122. Pfaller MA, Diekema DJ, Rex JH, Espinel -Ingroff A, Johnson EM, Andes D, et al. Correlation of MIC with outcome for Candi da species tested against voriconazole: analysis and proposal for inte rpretive breakpoints. Journal of clinical microbiology. 2006 Mar;44(3):819-26. 123. NCCLS. Reference method for broth dilution antifungal suscep tibility testing of yeasts. Approved standard, 2nd ed. NCCLS document M 27-A2. National Committee for Clinical Laboratory Standards, Wayne, PA.; 2002. 124. Smith J, Safdar N, Knasinski V, Simm ons W, Bhavnani SM, Ambrose PG, et al. Voriconazole therapeutic drug monitoring. Antimicrobial agents and chemotherapy. 2006 Apr;50(4):1570-2. 125. Liu P, Rand KH, Obermann B, Dere ndorf H. Pharmacokinetic-pharmacodynamic modelling of antibacterial ac tivity of cefpodoxime and cefixime in in vitro kinetic models. International jour nal of antimicrobial agents. 2005 Feb;25(2):120-9. 126. Schuck EL, Derendorf H. Pharmacokinetic/ pharmacodynamic evaluation of anti-infective agents. Expert review of anti-inf ective therapy. 2005 Jun;3(3):361-73. 127. Treyaprasert W, Schmidt S, Rand KH, Suvanakoot U, Derendorf H. Pharmacokinetic/pharmacodynamic modeling of in vitro activity of azithromycin against four different bacterial stra ins. International journal of antimicrobial agents. 2007 Mar;29(3):263-70. 128. Canton E, Peman J, Gobernado M, Viudes A, Espinel-Ingroff A. Syne rgistic activities of fluconazole and voriconazole with terbinafin e against four Candida species determined by checkerboard, time-kill, and Etest methods Antimicrobial agents and chemotherapy. 2005 Apr;49(4):1593-6. 129. Manavathu EK, Cutright JL, Chandrasekar PH. Organism-dep endent fungicidal activities of azoles. Antimicrobial agents a nd chemotherapy. 1998 Nov;42(11):3018-21. 130. Simonsen L, Jorgensen A, Benfeldt E, Groth L. Differentiated in vi vo skin penetration of salicylic compounds in hairle ss rats measured by cutaneous microdialysis. Eur J Pharm Sci. 2004 Feb;21(2-3):379-88. 131. Sasongko L, Williams KM, Day RO, McL achlan AJ. Human subcutaneous tissue distribution of fluconazole: co mparison of microdialysis and suction blister techniques. British journal of clinical pha rmacology. 2003 Nov;56(5):551-61. 132. Fair W, Jr. An Algorithm for Weight ed Linear Regressi on. Available from: http://www.codeproject.com/KB/recipes/LinReg.aspx 133. Araujo BV, Silva CF, Haas SE, Dalla Costa T. Microdialysis as a tool to determine free kidney levels of voriconazole in rodents: A model to study the technique feasibility for a moderately lipophilic drug. J ournal of pharmaceutical and biomedical analysis. 2008 Feb 29.

PAGE 152

152 134. Kusbeci T, Avci B, Cetinkaya Z, Ozturk F, Yavas G, Ermis SS, et al. The effects of caspofungin and voriconazole in experiment al Candida endophthalmitis. Current eye research. 2007 Jan;32(1):57-64. 135. Breit SM, Hariprasad SM, Mieler WF, Sh ah GK, Mills MD, Grand MG. Management of endogenous fungal endophthalmitis with voric onazole and caspofungin. American journal of ophthalmology. 2005 Jan;139(1):135-40. 136. Verweij PE, de Pauw BE, Meis JF. Voric onazole (Pfizer ltd). ID rugs. 1999 Sep;2(9):92537. 137. Yang H, Wang Q, Elmquist WF. The de sign and validation of a novel intravenous microdialysis probe: application to fluconazo le pharmacokinetics in the freely-moving rat model. Pharmaceutical research. 1997 Oct;14(10):1455-60. 138. Tsai TH. Assaying protein unbound drugs usi ng microdialysis tech niques. Journal of chromatography. 2003 N ov 25;797(1-2):161-73. 139. Liu Z, Yang XG, Li X, Pan W, Li J. Study on the ocular pharmacokinetics of ionactivated in situ gelling opht halmic delivery system for ga tifloxacin by microdialysis. Drug development and industrial pharmacy. 2007 Dec;33(12):1327-31. 140. Uehara T, Sumiyoshi T, Itoh H, Kurata K. Lactate production and neurotransmitters; evidence from microdialysis studies. Phar macology, biochemistry and behavior. 2008 Aug;90(2):273-81. 141. Sumiyoshi T, Tsunoda M, Ueha ra T, Tanaka K, Itoh H, Su miyoshi C, et al. Enhanced locomotor activity in rats with excitot oxic lesions of the entorhinal cortex, a neurodevelopmental animal model of schizophr enia: behavioral and in vivo microdialysis studies. Neuroscience letter s. 2004 Jul 1;364(2):124-9. 142. Apparaju SK, Gudelsky GA, Desai PB. Pharmacokinetics of gemcitabine in tumor and non-tumor extracellular fluid of brain: an in vivo assessment in rats employing intracerebral microdialysis. Cancer chemotherapy and pharmacology. 2008 Feb;61(2):223-9. 143. Zhou Q, Gallo JM. In vivo microdialysis for PK and PD studies of anticancer drugs. The AAPS journal. 2005;7(3):E659-67. 144. Schwarz AJ, Zocchi A, Reese T, Gozzi A, Garzotti M, Varnier G, et al. Concurrent pharmacological MRI and in situ microdialysis of cocaine reveal a complex relationship between the central hemodynamic respons e and local dopamine concentration. NeuroImage. 2004 Sep;23(1):296-304. 145. Egashira N, Ishigami N, Mishima K, Iwasaki K, Oishi R, Fujiwara M. Delta9Tetrahydrocannabinol-induced cognitive defi cits are reversed by olanzapine but not haloperidol in rats. Progre ss in neuro-psychopharmacology & biological psychiatry. 2008 Feb 15;32(2):499-506.

PAGE 153

153 146. van Duuren E, van der Plasse G, van der Blom R, Joosten RN, Mulder AB, Pennartz CM, et al. Pharmacological ma nipulation of neuronal ense mble activity by reverse microdialysis in freely moving rats: a compar ative study of the eff ects of tetrodotoxin, lidocaine, and muscimol. The Journal of pha rmacology and experimental therapeutics. 2007 Oct;323(1):61-9. 147. Nakashima K, Yamamoto K, Al-Dirba shi OY, Kaddoumi A, Nakashima MN. Semimicro column HPLC of triazolam in rat plasma and brain microdialysate and its application to drug interacti on study with itraconazole. Jo urnal of pharmaceutical and biomedical analysis. 20 03 Jan 15;30(6):1809-16. 148. Mathy FX, Ntivunwa D, Verbeeck RK, Preat V. Fluconazole distribution in rat dermis following intravenous and topi cal application: a microdi alysis study. Journal of pharmaceutical sciences. 2005 Apr;94(4):770-80. 149. Kondoh S, Nagasawa S, Kawanishi M, Yamaguc hi K, Kajimoto S, Ohta T. Effects of ebselen on cerebral ischemia and reperfus ion evaluated by microdi alysis. Neurological research. 1999 Oct;21(7):682-6. 150. Stimpfel TM, Gershey EL. Selecting anesth etic agents for human safety and animal recovery surgery. Faseb J. 1991 Apr;5(7):2099-104. 151. Martenson M, Houts J, Heinricher M, Ogden B. A simple device for humidification of inspired gases during volatile anesthesia in rats. Contemporary topics in laboratory animal science / American Association for Laboratory Animal Science. 2005 Mar;44(2):46-8. 152. Yang R, Tan X, Kenney RJ, Jr., Thomas A, Landis M, Qureshi N, et al. Hemorrhagic shock in the rat: comparison of carotid and femoral cannulation. The Journal of surgical research. 2008 Jan;144(1):124-6. 153. Beaumont DM, Mark TM, Jr., Hills R, Dixon P, Veit B, Garrett N. Th e effects of chrysin, a Passiflora incarnata extract, on natural kill er cell activity in male Sprague-Dawley rats undergoing abdominal surgery. AANA journal. 2008 Apr;76(2):113-7. 154. Araujo BV, Conrado DJ, Palma EC, Dalla Co sta T. Validation of rapid and simple LCMS/MS method for determination of voric onazole in rat plasma. Journal of pharmaceutical and biomedical analysis. 2007 Aug 15;44(4):985-90. 155. Muller M, dela Pena A, Derendorf H. Issues in pharmacokinetics and pharmacodynamics of anti-infective agents: distribution in tissu e. Antimicrobial agents and chemotherapy. 2004 May;48(5):1441-53. 156. Tegeder I, Lotsch J, Kinzig-Schippers M, Sorgel F, Kelm GR Meller ST, et al. Comparison of tissue concentrations after in tramuscular and topical administration of ketoprofen. Pharmaceutical research. 2001 Jul;18(7):980-6.

PAGE 154

154 157. Muller M, Rohde B, Kovar A, Georgopoulos A, Eichler HG, Derendorf H. Relationship between serum and free interstiti al concentrations of cefodizi me and cefpirome in muscle and subcutaneous adipose tissue of healthy volunteers measured by microdialysis. Journal of clinical pharmaco logy. 1997 Dec;37(12):1108-13.

PAGE 155

155 BIOGRAPHICAL SKETCH Yanjun Li was born in 1974, in Hunan, Ch ina. She got her M.D. in Hunan Medical University (current name: Central S outh Univer sity Xiangya School of Medicine). She started her Ph.D. program in August 2004 in the Department of Pharmaceutics, College of Pharmacy, University of Florida, under the supervision of Dr. Hartmut Derendorf. Yanjun received her Ph.D. in pharmaceutics in August 2008.